METHODS OF MODIFYING SEARCH AREAS
A video encoding method includes receiving a video captured by an image capture device on-board a movable object, where the video includes a video frame component; receiving sensor data from a plurality of sensors on-board the movable object; generating, according to the sensor data, an optical flow field associated with the video frame component; and evaluating motion of the video for video encoding based on the optical flow field.
This application is a continuation of application Ser. No. 15/452,680, filed on Mar. 7, 2017, which is a continuation of International Application No. PCT/CN2015/085762, filed on Jul. 31, 2015, the entire contents of both of which are incorporated herein by reference.
BACKGROUNDAerial vehicles, such as unmanned aerial vehicles (UAVs), have been developed for a wide range of applications including surveillance, search and rescue operations, exploration, and other fields. Such UAVs may often carry a camera module on-board for video capturing.
The video captured by UAVs may be encoded by various methods. However, video encoding methods and systems for UAVs may be less than ideal. For example, packet loss may occur when captured video from a UAV is encoded and transmitted, especially when the video contains a large amount of movement.
SUMMARYA need exists for improved methods and systems for encoding video obtained from video capture devices. The video capture devices may be carried by unmanned vehicles, such as unmanned aerial vehicles (UAVs). Methods are provided for encoding video captured by video capture devices, such as video capture devices on UAVs, by utilizing information from sensors associated with the UAVs. In some embodiments, the video capture devices may capture video that includes motion data. Additionally, a UAV may use sensors that are associated with the UAV to capture information that may be used to generate an optical flow field. When the captured video is aligned with a correlating optical flow field that is based on sensor information captured at a similar time as the video, the resulting information may be used to efficiently encode the video data. In particular, the aligned video and optical flow field data may be used to efficiently determine the location of motion vectors within the video. The motion vectors, in turn, may be used to reduce residual data that results from differences between a predicted video frame and a perceived video frame during processes of video encoding. Accordingly, methods directed towards accurately and efficiently identifying motion vectors may increase the efficiency and usefulness of video encoding processes.
An optical flow field that is generated using sensor data from a UAV may be used to efficiently encode video data that is aligned with the generated optical flow field. The video data may be encoded by a processor at the UAV, video capture device, or carrier on-board the UAV. The video data may be encoded by a processor external to the UAV, such as a user terminal that is communicatively connected to the UAV. Additionally, the optical flow field may be generated at the UAV. Alternatively, the optical flow field may be generated at an external location that is communicatively connected to the UAV. The sensor information that is used to generate the optical flow field may be detected at the UAV. Additionally or alternatively, the sensor information that is used to generate the optical flow field may be provided to the UAV from an external source that is communicatively connected to the UAV. Accordingly, video data that is captured by a video capture device may be efficiently encoded using an optical flow field that is generated based on sensor data that is associated with the UAV.
In particular, an optical flow field that corresponds to video data captured by a UAV may be used to efficiently identify a location of a motion vector associated with a video frame component within the video data. When encoding video data, it is beneficial to break up video data into video frame components and encode recognized similarities between video frame components, rather than encoding each frame over and over again. However, as some video data may have a great deal of movement, it is sometimes difficult to associate blocks of a first video frame with blocks of a second video frame, even when there may be a great amount of similarity between the two video frames. This is because, with great movement, the bias of the similar elements within a video frame may be shifted across a video frame. For example, as a camera shifts right, objects of the video that were formerly at the right edge of a video frame will be shifted to the left. However, conventional methods of encoding video data are based on the assumption that blocks at a particular location on a first video frame are associated with blocks at the same particular location on a second video frame. While conventional methods do make use of a search area so as to identify areas of interest surrounding an initial, co-located point of the block on the second video frame, the size and location of the search area that is used is not changed in view of movement of the camera. As such, the use of conventional methods to find connections, such as identifying shared motion vectors, between similar blocks among different video frames are generally lacking when there is a significant amount of movement associated with the video data.
Based on this shortcoming of conventional methods of encoding video data, aspects of the disclosure provide the use of optical flow field data to contextualize video data. In particular, an optical flow field that is aligned with the video data may be used to modify the size and shape of search areas used to identify connecting information between blocks across video frames, such as motion vectors. Additionally, the optical flow field may also be used to bias/shift an initial starting point of a search area for a motion vector associated with a block of a video frame so as to more accurately identify the motion vector within the block.
An aspect of the disclosure may include a method of evaluating a search area for encoding video. The method may comprise receiving video captured by an image capture device, the video comprising video frame components. Additionally, the method may comprise receiving optical flow field data associated with the video frame component, wherein at least a portion of the optical flow field data is captured by sensors. The method may also comprise determining a search area based on the optical flow field data.
Additionally, aspects of the disclosure may include a non-transitory computer readable medium containing program instructions for evaluating a search area for encoding video. The computer readable medium may include program instructions for receiving video captured by an image capture device, the video comprising video frame components. The computer readable medium may also include program instructions for receiving optical flow field data associated with the video frame component, wherein at least a portion of the optical flow field data is captured by sensors. Further, the computer readable medium may include program instructions for determining a search area based on the optical flow field data.
Aspects of the disclosure may also include a system for evaluating a search area for encoding video. The system may include an image capture device configured to capture a video. Additionally, the system may include one or more sensors configured to capture optical flow field data. The system may also include one or more processors. The one or more processors may be individually or collectively configured to receive the video captured by the image capture device. The video may comprise video frame components. Additionally, the one or more processors may be configured to receive optical flow field data associated with the video frame component, wherein at least a portion of the optical flow field data is captured by the one or more sensors. The one or more processors may also be configured to determine a search area based on the optical flow field data.
Aspects of the disclosure may further include a method of identifying a motion vector of a video frame component. The method may comprise receiving video captured by an image capture device, the video comprising the video frame component. Additionally, the method may comprise receiving optical flow field data associated with the video frame component. The method may also comprise determining a search area for assessing a motion vector, wherein the search area is based on the received optical flow field data.
Additionally, aspects of the disclosure may include a non-transitory computer readable medium containing program instructions for determining a motion vector of a video frame component. The computer readable medium may include program instructions for receiving video captured by an image capture device, the video comprising the video frame component. The computer readable medium may also include program instructions for receiving optical flow field data associated with the video frame component. Additionally, the computer readable medium may include program instructions for determining a search area for assessing the motion vector, wherein the search area is based on the received optical flow field data.
Aspects of the disclosure may also include a system for identifying a motion vector of a video frame component. The system may include an image capture device configured to capture a video. The system may also include one or more sensors configured to capture optical flow field data. Additionally, the system may include one or more processors. The one or more processors may be configured to receive video captured by the image capture device, the video comprising the video frame component. The one or more processors may also be configured to receive the optical flow field data associated with the video frame component. Additionally, the one or more processors may be configured to determine a search area for assessing the motion vector, the search area comprising the initial point, wherein the dimensions of the search area is based on the received optical flow field data.
In further aspects of the disclosure may include a method of determining a search area for encoding video. The method may comprise receiving video captured by an image capture device, the video comprising a video frame. The method may also comprise receiving motion data associated with the video frame, wherein the motion data are captured by sensors. Additionally, the method may comprise determining a search area within the video frame based on the motion data, wherein the search area is used to evaluate motion of the video captured by the image capture device.
Additional aspects of the disclosure may include a non-transitory computer readable medium containing program instructions for determining a search area for encoding video. The computer readable medium may include program instructions for receiving video captured by an image capture device, the video comprising a video frame. Additionally, the computer readable medium may include program instructions for receiving motion data associated with the video frame, wherein the motion data are captured by sensors, and wherein the video frame comprises a plurality of blocks. Further, the computer readable medium may include program instructions for determining a search area within the video frame based on the motion data.
Aspects of the disclosure may also include a system for determining a search area for encoding video. The system may include an image capture device configured to capture a video. Additionally, the system may include one or more sensors configured to capture motion data. The system may also include one or more processors. The one or more processors may be configured to receive the video captured by the image capture device, the video comprising a video frame. The one or more processors may also be configured to receive motion data associated with the video frame, wherein the motion data are captured by sensors, and wherein the video frame comprises a plurality of blocks. Additionally, the one or more processors may be configured to determine a search area within the video frame based on the motion data.
Additional aspects of the disclosure may include a method of determining a search area for encoding video. The method may include receiving video captured by an image capture device, the video comprising a first video frame and a second video frame, each of which comprises a plurality of blocks. The method may also include receiving motion data associated with the second video frame, wherein the motion data are captured by sensors. Additionally, the method may include determining a search area within the first video frame based on the motion data associated with the second video frame.
Additional aspects of the disclosure may include a non-transitory computer readable medium containing program instructions for determining a search area for encoding video. The computer readable medium may include program instructions for receiving video captured by an image capture device, the video comprising a first video frame and a second video frame, each of which comprises a plurality of blocks. Additionally, the computer readable medium may include program instructions for receiving motion data associated with the second video frame, wherein the motion data are captured by sensors. Further, the computer readable medium may include program instructions for determining a search area within the first video frame based on the motion data associated with the second video frame.
Aspects of the disclosure may also include a system for determining a search area for encoding video. The system may include an image capture device configured to capture a video. Additionally, the system may include one or more sensors configured to capture motion data. The system may also include one or more processors. The one or more processors may be configured to receive the video captured by the image capture device, the video comprising a first video frame and a second video frame, each of which comprises a plurality of blocks. The one or more processors may also be configured to receive motion data associated with the second video frame, wherein the motion data are captured by sensors. Additionally, the one or more processors may be configured to determine a search area within the first video frame based on the motion data associated with the second video frame.
Another aspect of the disclosure may include a method of encoding video captured by an aerial vehicle. The method may comprise receiving video captured by an image capture device on-board the aerial vehicle, the video comprising video frame components. Additionally, the method may comprise receiving optical flow field data associated with a video frame component, wherein at least a portion of the optical flow field data is captured by sensors on-board the aerial vehicle. The method may also comprise evaluating motion of the video captured by the image capture device for video encoding, based on the optical flow field data.
Additionally, aspects of the disclosure may include a non-transitory computer readable medium containing program instructions for encoding video captured by an aerial vehicle. The computer readable medium may include program instructions for receiving video captured by an image capture device on-board the aerial vehicle, the video comprising video frame components. The computer readable medium may also include program instructions for receiving optical flow field data associated with a video frame component, wherein at least a portion of the optical flow field data is captured by sensors on-board the aerial vehicle. Further, the computer readable medium may include program instructions for evaluating motion of the video captured by the image capture device for video encoding, based on the optical flow field data.
Aspects of the disclosure may also include a system for encoding video captured by an aerial vehicle. The system may include an image capture device on-board an aerial vehicle, said image capture device configured to capture a video. Additionally, the system may include one or more sensors on-board the aerial vehicle, said one or more sensors configured to capture optical flow field data. The system may also include one or more processors. The one or more processors may be individually or collectively configured to receive the video captured by an image capture device, the video comprising video frame components. Additionally, the one or more processors may be configured to receive the optical flow field data associated with a video frame component, wherein at least a portion of the optical flow field data is captured by the sensors. The one or more processors may also be configured to evaluate motion of the video captured by the image capture device for video encoding, based on the optical flow field data.
It shall be understood that different aspects of the disclosure may be appreciated individually, collectively, or in combination with each other. Various aspects of the disclosure described herein may be applied to any of the particular applications set forth below or for any other types of movable objects. Any description herein of aerial vehicles, such as unmanned aerial vehicles, may apply to and be used for any movable object, such as any vehicle. Additionally, the systems, devices, and methods disclosed herein in the context of encoding video while a video capture device is capturing video data of aerial motion (e.g., flight) may also be applied in the context of encoding video while a video capture device is capturing video data of other types of motion, such as movement on the ground or on water, underwater motion, or motion in space.
Other objects and features of the present disclosure will become apparent by a review of the specification, claims, and appended figures.
INCORPORATION BY REFERENCEAll publications, patents, and patent applications mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference.
The novel features of the invention are set forth with particularity in the appended claims. A better understanding of the features and advantages of the present disclosure will be obtained by reference to the following detailed description that sets forth illustrative embodiments, in which the principles of the disclosure are utilized, and the accompanying drawings of which:
The methods, devices and terminals described herein provide effective approaches for efficiently encoding video captured by a video capture device, such as video capture devices on-board unmanned aerial vehicles (UAVs). The methods, devices and terminals described herein may be used to capture video data, generate an optical flow field based on sensor data associated with the UAV, and modify search areas for use in encoding the video data based on the generated optical flow field. The methods, devices and terminals disclosed herein may be applied to video that is captured by a video capture device that is attached to any suitable movable object or stationery objects. A movable object may be capable of self-propelled movement (e.g., a vehicle), while a stationary object may not be capable of self-propelled movement. In some embodiments, the movable object may be an unmanned aerial vehicle (UAV).
Accordingly, methods that are provided may be used to efficiently encode video data. When video data is more efficiently encoded, a larger amount of video data may be recorded given a set amount of storage space. Alternatively, video that has increased capacity may be recorded within the same amount of storage space that previously would only be able to record the same amount of general video data. This is beneficial in recording high-definition video, or in recording video having a high degree of movement.
The way methods of the disclosure are able to efficiently encode video data is by efficiently searching for motion vectors that link aspects of video frames across time. In particular, the methods for searching for these linking aspects are contextualized based on other influences that are affecting objects that are being captured in the video data, such as influences associated with movement of the video capture device. In this way, by contextualizing motion vectors with the movement of the video capture device and other influences that are captured by sensors of a UAV that is used to mount the video capture device, the search area needed to identify these linking motion vectors is minimized. As such, the effort needed to encode the video data is minimized.
Video may be captured using a video capture device 140. The video capture device may be supported on a stationary object or a movable object, such as a UAV. Any description herein of a UAV may include any support structure for the video capture device. Any description herein of a UAV 100 may apply to any type of movable object, such as an aerial vehicle. The description of a UAV may apply to any type of unmanned movable object (e.g., which may traverse the air, land, water, or space). The UAV may be capable of responding to commands from a remote controller. The remote controller may be not connected to the UAV; rather, the remote controller may communicate with the UAV wirelessly from a distance. In some instances, the UAV may be capable of operating autonomously or semi-autonomously. The UAV may be capable of following a set of pre-programmed instructions. In some instances, the UAV may operate semi-autonomously by responding to one or more commands from a remote controller while otherwise operating autonomously. For instance, one or more commands from a remote controller may initiate a sequence of autonomous or semi-autonomous actions by the UAV in accordance with one or more parameters. In some embodiments, any description herein of a UAV may apply to any stationary object, such as a support for the video capture device (e.g., stand, pole, fence, building, wall, ceiling, roof, floor, ground, furniture, lighting fixture, tree, plant, stone, or any other stationary object).
The video capture device may be capable of altering a field of view (FOV) captured by the video capture device. The video capture device may have translational motion (e.g., side to side, front to back, up and down, or any combination thereof) to alter the video capture device FOV. The video capture device may have rotational movement (e.g., about a yaw, pitch, or roll axis of the video capture device) to alter the video capture device FOV. In some instances, the video capture device may only have translational motion without rotational motion, may only have rotational motion without translational motion, or may have both translational and rotational motion. Motion captured by video from the video capture device may be indicative of change of the video capture device FOV. The video encoding systems and methods may be used to encode the video captured by the video capture device, as described in greater detail elsewhere herein.
The video capture device may optionally be supported by a UAV 100 or any other support structure. The UAV may have a body 110. In some instances, the body may be a central body which may have one or more branching members, or “arms.” The arms may extend outward from the body in a radial manner and be joined via the body. The number of arms may match the number of propulsion units, or rotors, of the UAV. The body may comprise a housing. The housing may enclose one or more components of the UAV within the housing. In some instances, one or more electrical components of the UAV may be provided within the housing. For example, a flight controller of the UAV may be provided within the housing. The flight controller may control operation of one or more propulsion units 120 of the UAV. The propulsion units may each include the rotors and/or motors. Additionally, the one or more propulsion units may permit the UAV to move about in the air. The one or more propulsion units may be provided on an arm of the UAV. The arm may be connected to a body of the UAV on a proximal end of the arm. One or more propulsion units may be connected to a distal end of the arm. The one or more propulsion units may enable the UAV to move about one or more, two or more, three or more, four or more, five or more, six or more degrees of freedom. In some instances, the UAV may be able to rotate about one, two, three or more axes of rotation. The axes of rotation may be orthogonal to one another. The axes of rotation may remain orthogonal to one another throughout the course of the UAV's flight. The axes of rotation may include a pitch axis, roll axis, and/or yaw axis. The UAV may be able to move along one or more dimensions. For example, the UAV may be able to move upwards due to the lift generated by one or more rotors. In some instances, the UAV may be capable of moving along a Z axis (which may be up relative to the UAV orientation), an X axis, and/or a Y axis (which may be lateral). The UAV may be capable of moving along one, two, or three axes that may be orthogonal to one another.
The UAV may be a rotorcraft. In some instances, the UAV may be a multi-rotor craft that may include a plurality of rotors. The plurality of rotors may be capable of rotating to generate lift for the UAV. The rotors may be propulsion units that may enable the UAV to move about freely through the air. The rotors may rotate at the same rate and/or may generate the same amount of lift or thrust. The rotors may optionally rotate at varying rates, which may generate different amounts of lift or thrust and/or permit the UAV to rotate. In some instances, one, two, three, four, five, six, seven, eight, nine, ten, or more rotors may be provided on a UAV. The rotors may be arranged so that their axes of rotation are parallel to one another. In some instances, the rotors may have axes of rotation that are at any angle relative to one another, which may affect the motion of the UAV.
The UAV shown may have a plurality of rotors. The rotors may connect to the body of the UAV which may comprise a control unit, one or more sensors, a processor, and a power source. The sensors may include vision sensors and/or other sensors that may collect information about the UAV environment. The information from the sensors may be used to determine a location of the UAV. The rotors may be connected to the body via one or more arms or extensions that may branch from a central portion of the body. For example, one or more arms may extend radially from a central body of the UAV, and may have rotors at or near the ends of the arms.
A vertical position and/or velocity of the UAV may be controlled by maintaining and/or adjusting output to one or more propulsion units of the UAV. For example, increasing the speed of rotation of one or more rotors of the UAV may aid in causing the UAV to increase in altitude or increase in altitude at a faster rate. Increasing the speed of rotation of the one or more rotors may increase the thrust of the rotors. Decreasing the speed of rotation of one or more rotors of the UAV may aid in causing the UAV to decrease in altitude or decrease in altitude at a faster rate. Decreasing the speed of rotation of the one or more rotors may decrease the thrust of the one or more rotors. When a UAV is taking off, the output provided to the propulsion units may be increased from its previous landed state. When the UAV is landing, the output provided to the propulsion units may be decreased from its previous flight state. The UAV may be configured to take off and/or land in a substantially vertical manner.
A lateral position and/or velocity of the UAV may be controlled by maintaining and/or adjusting output to one or more propulsion units of the UAV. The altitude of the UAV and the speed of rotation of one or more rotors of the UAV may affect the lateral movement of the UAV. For example, the UAV may be tilted in a particular direction to move in that direction and the speed of the rotors of the UAV may affect the speed of the lateral movement and/or trajectory of movement. Lateral position and/or velocity of the UAV may be controlled by varying or maintaining the speed of rotation of one or more rotors of the UAV.
The arms of the UAV may be tubes or rods. The arms of the UAV may have a circular cross section. The arms of the UAV may have a square or rectangular cross section. The arms of the UAV may have an elliptic cross section. The arms of the UAV may be hollow tubes. The arms of the UAV may be solid tubes. The arms of the UAV may be formed from a metallic, plastic, or composite material. The arms of the UAV may be formed from a lightweight material. The arms of the UAV may be formed from carbon fiber. The arms of the UAV may be integrally formed with the central body of the UAV. Alternatively, the arms of the UAV may be separately formed or may be separable from the UAV.
The UAV may have a greatest dimension (e.g., length, width, height, diagonal, diameter) of no more than 100 cm. In some instances, the greatest dimension may be less than or equal to 1 mm, 5 mm, 1 cm, 3 cm, 5 cm, 10 cm, 12 cm, 15 cm, 20 cm, 25 cm, 30 cm, 35 cm, 40 cm, 45 cm, 50 cm, 55 cm, 60 cm, 65 cm, 70 cm, 75 cm, 80 cm, 85 cm, 90 cm, 95 cm, 100 cm, 110 cm, 120 cm, 130 cm, 140 cm, 150 cm, 160 cm, 170 cm, 180 cm, 190 cm, 200 cm, 220 cm, 250 cm, or 300 cm. Optionally, the greatest dimension of the UAV may be greater than or equal to any of the values described herein. The UAV may have a greatest dimension falling within a range between any two of the values described herein. The UAV may be lightweight UAV. For example, the UAV may weigh less than or equal to 1 mg, 5 mg, 10 mg, 50 mg, 100 mg, 500 mg, 1 g, 2 g, 3 g, 5 g, 7 g, 10 g, 12 g, 15 g, 20 g, 25 g, 30 g, 35 g, 40 g, 45 g, 50 g, 60 g, 70 g, 80 g, 90 g, 100 g, 120 g, 150 g, 200 g, 250 g, 300 g, 350 g, 400 g, 450 g, 500 g, 600 g, 700 g, 800 g, 900 g, 1 kg, 1.1 kg, 1.2 kg, 1.3 kg, 1.4 kg, 1.5 kg, 1.7 kg, 2 kg, 2.2 kg, 2.5 kg, 3 kg, 3.5 kg, 4 kg, 4.5 kg, 5 kg, 5.5 kg, 6 kg, 6.5 kg, 7 kg, 7.5 kg, 8 kg, 8.5 kg, 9 kg, 9.5 kg, 10 kg, 11 kg, 12 kg, 13 kg, 14 kg, 15 kg, 17 kg, or 20 kg. The UAV may have a weight greater than or equal to any of the values described herein. The UAV may have a weight falling within a range between any two of the values described herein.
The UAV may carry the video capture device 140. The video capture device may be supported by any support structure, moving (e.g., UAV) or stationary. In some embodiments, the video capture device may be a camera. Any description herein of a camera may apply to any type of video capture device. The camera may be rigidly coupled to the support structure. Alternatively, the camera may be permitted to move relative to the support structure with respect to up to six degrees of freedom. The camera may be directly mounted onto the support structure, or coupled to a carrier mounted onto the support structure. In some embodiments, the carrier may be a gimbal. In some embodiments, the camera may be an element of a payload of the support structure, such as a UAV.
The camera may capture images (e.g., dynamic images such as video, or still images such as snapshots) of an environment of the UAV. The camera may continuously capture images (e.g., video). Alternatively, the camera may capture images (e.g., video) at a specified frequency to produce a series of image data (e.g., video data) over time. Any description herein of video may apply to any type of images, such as dynamic or still images, such as a series of images captured over time. Images may be captured at a video rate (e.g., 25, 50, 75, 100, 125, 150, 175, 200, 225, or 250 Hz). In some embodiments, the video may be captured simultaneously with a recording of environment audio.
In some embodiments, the captured video may be stored in a memory on-board the UAV. The memory may be a non-transitory computer readable medium that may include one or more memory units (e.g., removable media or external storage such as a Secure Digital (SD) card, or a random access memory (RAM), or a read only memory (ROM) or a flash memory). Alternatively, the captured video and/or images may be transmitted to a remote terminal. The transmission of captured video and/or images may be implemented over a wireless link, including but not limited to, a radio frequency (RF) link, a Wi-Fi link, a blue tooth link, a 2G link, a 3G link, or a LTE link. The memory may be on the camera carried by the UAV, on a carrier of the UAV, and/or on the UAV itself (e.g., within the UAV body or an arm of the UAV). The memory may or may not be removable or separable from the UAV, carrier, or camera.
The camera may comprise an image sensor and one or more lenses. The one or more lenses may be configured to direct light to the image sensor. An image sensor is a device that converts an optical image into an electronic signal. The image sensor of the camera may be a charge-coupled device (CCD) type, a complementary metal-oxide-semiconductor (CMOS) type, an N-type metal-oxide-semiconductor (NMOS) type, or a back-side illuminated CMOS (BSI-CMOS) type.
The camera may have a focal length or focal length range. A focal length of an optical system may be a measure of how strongly the system converges or diverges light. The focal length that is associated with the camera may influence a resulting optical flow field that is generated using video that is captured by the camera. The focal length of a lens may be the distance over which initially collimated rays are brought to a focus. The camera may have any type of lens, such as a prime lens or a zoom lens. A prime lens may have a fixed focal length and the focal length may encompass a single focal length. A zoom lens may have variable focal lengths and the focal length may encompass a plurality of focal lengths.
The video capture device may have a FOV that may change over time. The field of view (FOV) may be a part of the world that is visible through the camera at a particular position and orientation in space; objects outside the FOV when the picture is taken are not recorded in the video data. It is most often expressed as the angular size of the view cone, as an angle of view. For normal lens, field of view may be calculated as FOV=2 arctan(d/2f), where d is image sensor size, and f is focal length of the lens. For an image sensor having a fixed size, the prime lens may have a fixed FOV and the FOV may encompass a single FOV angle. For an image sensor having a fixed size, the zoom lens may have variable FOV angular range and the FOV angular range may encompass a plurality of FOV angles. The size and/or location of the FOV may change. The FOV of the video capture device may be altered to increase or decrease the size of the FOV (e.g., zooming in or out), and/or to change a centerpoint of the FOV (e.g., moving the video capture device translationally and/or rotationally). Alteration of the FOV may result in motion within the video.
Data from sensors associated with a camera may be used to aid in generating an optical flow field, useful for encoding video data captured by the camera. The sensors associated with the camera may be on-board the camera, the support structure for the camera (e.g., UAV), and/or a carrier that supports the camera on the support structure (e.g., gimbal). Alternatively, the sensors associated with the camera may be remote from the camera, the carrier, and/or the support structure for the camera.
For instance, a support structure of the camera may support one or more sensors. In examples, the support structure may be a UAV. Any description of the sensors of the UAV may apply to any type of support structure for the camera. The UAV may comprise one or more vision sensors such as an image sensor. For example, an image sensor may be a monocular camera, stereo vision camera, radar, sonar, or an infrared camera. The UAV may further comprise other sensors that may be used to determine a location of the UAV, or may be useful for generating optical flow field information, such as global positioning system (GPS) sensors, inertial sensors which may be used as part of or separately from an inertial measurement unit (IMU) (e.g., accelerometers, gyroscopes, magnetometers), lidar, ultrasonic sensors, acoustic sensors, WiFi sensors. The UAV may have sensor on-board the UAV that collect information directly from an environment without contacting an additional component off-board the UAV for additional information or processing. For example, a sensor that collects data directly in an environment may be a vision or audio sensor.
Alternatively, the UAV may have sensors that are on-board the UAV but contact one or more components off-board the UAV to collect data about an environment. For example, a sensor that contacts a component off-board the UAV to collect data about an environment may be a GPS sensor or another sensor that relies on connection to another device, such as a satellite, tower, router, server, or other external device. Various examples of sensors may include, but are not limited to, location sensors (e.g., global positioning system (GPS) sensors, mobile device transmitters enabling location triangulation), vision sensors (e.g., imaging devices capable of detecting visible, infrared, or ultraviolet light, such as cameras), proximity or range sensors (e.g., ultrasonic sensors, lidar, time-of-flight or depth cameras), inertial sensors (e.g., accelerometers, gyroscopes, inertial measurement units (IMUs)), altitude sensors, attitude sensors (e.g., compasses) pressure sensors (e.g., barometers), audio sensors (e.g., microphones) or field sensors (e.g., magnetometers, electromagnetic sensors). Any suitable number and combination of sensors may be used, such as one, two, three, four, five, or more sensors. Optionally, the data may be received from sensors of different types (e.g., two, three, four, five, or more types). Sensors of different types may measure different types of signals or information (e.g., position, orientation, velocity, acceleration, proximity, pressure, etc.) and/or utilize different types of measurement techniques to obtain data.
Any of these sensors may also be provided off-board the UAV. The sensors may be associated with the UAV. For instance, the sensors may detect characteristics of the UAV such as position of the UAV, speed of the UAV, acceleration of the UAV, orientation of the UAV, noise generated by the UAV, light emitted or reflected from the UAV, heat generated by the UAV, or any other characteristic of the UAV. The sensors may collect data that may be used alone or in combination with sensor data from sensors on-board the UAV to generate optical flow field information.
The sensors may include any suitable combination of active sensors (e.g., sensors that generate and measure energy from their own energy source) and passive sensors (e.g., sensors that detect available energy). As another example, some sensors may generate absolute measurement data that is provided in terms of a global coordinate system (e.g., position data provided by a GPS sensor, attitude data provided by a compass or magnetometer), while other sensors may generate relative measurement data that is provided in terms of a local coordinate system (e.g., relative angular velocity provided by a gyroscope; relative translational acceleration provided by an accelerometer; relative attitude information provided by a vision sensor; relative distance information provided by an ultrasonic sensor, lidar, or time-of-flight camera). The sensors on-board or off-board the UAV may collect information such as location of the UAV, location of other objects, orientation of the UAV 100, or environmental information. A single sensor may be able to collect a complete set of information in an environment or a group of sensors may work together to collect a complete set of information in an environment. Sensors may be used for mapping of a location, navigation between locations, detection of obstacles, or detection of a target. Additionally, and in accordance with the disclosure, the sensors may be used to gather data which is used to generate an optical flow field that is used to efficiently encode video data captured by the UAV.
Accordingly, the UAV may also have an optical flow field generator 130. The optical flow field generator may be provided on-board the UAV (e.g., in the UAV body or arm, on the camera, or on the carrier). Alternatively, the optical flow field generated may be provided off-board the UAV (e.g., at a remote server, cloud computing infrastructure, remote terminal, or ground station). The optical flow field generator may have one or more processors that are individually or collectively configured to generate an optical flow field based on sensor data that is associated with the UAV. An optical flow field demonstrates how light flows within video frames. This flow of light indicates how captured objects are moving between video frames. In particular, the optical flow field is able to describe characteristics of how objects that are captured by a video capturing device are moving. For instance, the video captured within the FOV of the video capturing device may include one or more stationary or movable objects. In examples, the optical flow field may be used to determine speeds or accelerations of objects that are moving in video. The optical flow field may also be used to demonstrate directions of movement of objects that are within the video. Examples of optical flow fields that describe objects moving within a video are described below with respect to
The sensor data that is used to generate the optical flow field may be obtained by the one or more sensors associated with the UAV. Additionally or alternatively, the sensor data may be obtained by an external source, such as an external monitoring system. The external sensor data may be provided to the UAV using a communication channel. Accordingly, the optical flow field may be generated at the UAV. Alternatively, an optical flow field may be generated external to the UAV. In particular, the UAV may provide sensor information that is associated with the UAV to one or more external processors. The one or more external processors may then use the sensor data that is associated with the UAV to generate an optical flow field. Further, the one or more external processors may provide the optical flow field that is generated to the UAV. The optical flow field generator, whether on-board or off-board the UAV, may receive data from sensors associated with the UAV (whether the sensors are on-board, off-board, or any combination thereof), which may be used to generate an optical flow field.
The sensor data may optionally include information about the spatial disposition of the camera (e.g., coordinates, translational position, height, orientation), or movement of the camera (e.g., linear speed, angular speed, linear acceleration, angular acceleration). The sensor data may be able to detect a zoom state of the camera (e.g., focal length, how far zoomed in or out). The sensor data may be useful for calculating how a FOV of the camera may change.
An encoding processor 150 may be provided in accordance with embodiments of the disclosure. The encoding processor may be used to encode video that is captured by the video capture device. Examples of entropy coding tools include Huffman coding, run-level coding, and arithmetic coding. In examples discussed herein, context-adaptive variable-length coding (CAVLC) and context-adaptive binary arithmetic coding (CABAC) may be used in H264.
Additionally, the encoding processor may use an optical flow field that is associated with the video. In particular, the optical flow field may be used to efficiently encode the video. The encoding processor may receive the optical flow field information and use the optical flow field information to encode the video.
The encoding processor may include one or more processors that may encode the video. The encoding processor may be separate from the optical flow field generator, or may be the within the same component as the optical flow field generator. The encoding processor may include one or more processors that do not overlap with one or more processors of the optical flow field generator. Alternatively, one or more processors of the encoding processor may be the same as one or more processors of the optical flow field generator. In some instances, all processors of the encoding processor may be the same as the processors of the optical flow field generator.
The encoding processor may optionally be provided on-board the UAV. For instance, the encoding processor may be within the UAV body or arm, may be on-board the camera, or may be on-board a carrier supporting the camera. Alternatively, the encoding processor may be provided off-board the UAV. For instance, the encoding processor may be provided at a remote server, cloud computing infrastructure, remote terminal, or ground station. The encoding processor may be provided at a same or different location from the optical flow field generator.
As seen in
A video encoder may be used to remove the correlation of the blocks spatially and/or temporally. As such, after a video frame is divided into small blocks, the blocks of video data may go through a video encoding architecture as provided in
In particular, the video data may proceed to a coder control 204. The coder control may be used to determine whether to encode the video data directly, e.g. without any additional transformation steps, or whether to send the data to a transformation/scaling/quantization (TSQ) component. In examples, the coder control may pass the video data directly to an entropy coding component 206. In other examples, the coder control may pass the video data to a TSQ component 208 prior to providing the transformed data to the entropy coding component. At the TSQ component, the video data may be transformed so as to compress similarities between spatially and temporally related video frame components, such as blocks. This process may use video from the original input video signal. Additionally, this process may utilize previously encoded video data so as to make the transformation process more efficient. Additionally, this compression process may result in quantization and transformation coefficients 210 which may then be provided to the entropy encoding component.
When transforming the video data, the video data may be processed in view of previously transformed video data that is re-evaluated at decoder 212 and that is provided as feedback to the TSQ component. In particular, video compression feedback may be generated by providing transformed video data from the TSQ component to scaling and inversion transformation (SIT) component 214. At the SIT component, the transformation process of the video data may be reversed. This video data may then be provided to a de-blocking filter 216 which may be used to generate an output video signal 218. The output video signal may then be used as a component to generate motion compensation factors at motion compensation component 220.
In examples, the motion compensation component may use motion data from an output video signal as well as motion data that is generated from motion estimation component 222. In particular, the motion estimation component may receive input video data from the initial input video signal. The motion estimation component may then generate motion data based on the video data. This motion data may then be provided to the motion compensation component and the entropy coding component.
Once the decoded video data is provided and contextualized based on motion data from the motion compensation component, the video data may be evaluated for intra frame prediction using intra-frame prediction component 224. Additional predictions may also be generated for inter-frame predictions. These predications may be provided as feedback for both the TSQ component as well as the de-blocking filter. As such, the quantization and transformation coefficients that are generated from the TSQ component, as well as the output signal that is generated by the de-blocking filter, may be refined based on feedback from processed video data.
As such, a video encoder may be used to simplify duplicate information, both between blocks of different video frames (temporal compression) as well as between blocks within the same video frame (spatial compression), so as to condense information. Once the video data is condensed, the video frames that are encoded utilizing the architecture in
Intra coding may be used to condense spatial correlations. For a block within a video frame, a predictor of pixel values within the block may be estimated from its neighboring pixels. For example, a predictor of pixel values may be estimated from neighboring pixels such as the upper, left upper right, and lower left neighboring pixels. Examples of these predications may be directional so as to correspond with the pattern within a pixel block. A demonstration of H.264 directional intra prediction is provided in
Additionally, the mode that is assigned to the adjacent pixels may be used to determine the predictive motion of the pixels in the block. For example, in mode 0, the pixels that are adjacent to a block may be assessed as having a downward motion. As seen in
In mode 1, the pixels that are adjacent to a block may be assessed as having a sideways motion. As seen in
In mode 2, the pixels that are adjacent to the block may be assessed as having a normal, or neutral, movement. Based on this assessment, the pixels within the block may be assessed to have a neutral movement as well. In examples, pixels within the block may be assessed to have a neutral movement when the objects that are captured within the video capture device are not moving; when objects that are captured within the video capture device do not appear to moving because the movement of an object is equal to that of a moving image capture device; or when objects that are captured within the video capture device do not appear to be moving because the objects are so far away that incremental movements made in relation to the far-away objects appear to be against a backdrop that is relatively still.
In mode 3, the pixels that are adjacent to a block, and in close proximity to the upper portion of the block, may be assessed as having a leftward angled motion. As seen in
In mode 4, the pixels that are adjacent to the block may be assessed as having a rightward angled motion. As seen in
Additionally, mode 8 provides adjacent pixels to a block that indicate a motion that is upwards and to the right. However, mode 8 differs from previous modes in that mode 8 is only able to predict a portion of the block. For assessing the additional predictive pixels within the block, other auxiliary methods may be used.
While intra coding utilizes neighboring pixels of a block, such as pixels on the left column and the upper row of a current block, there may be a significant amount of residual information that is included within the central pixels of a block. In examples, the central pixels of a block may include textures, objects, and other information that may not be readily predicted using intra coding. To capture this information, information between frames (e.g. temporal compression) may be condensed and encoded.
Inter coding may be used to condense temporal correlations. For a block within a video frame, a predictor of pixel values within the block may be estimated from a correlating block within a previous frame. As video frames may only be separated by a few millionths of a second, blocks between frames may not generally differ greatly. However, the use of inter coding may be useful for predicting details within a block that would not be captured using intra frame coding. In particular, these details are predicted by referencing block from nearby video frames. In particular, blocks that are correlated between frames may be linked using a motion vector.
When implementing inter coding, initially an inter frame motion estimation may be performed on the encoding block. The motion estimation process may determine a grid of pixels which may be considered most similar and most costless to a current block. In particular, the motion estimation may determine the grid of pixels that is considered most similar by conducting a search within a search area of a video frame. Once a grid of pixels which is considered the most similar and most costless to the current block is determined, a motion vector may be calculated. In particular, the motion vector may be calculated as comprising the 2D pixel location difference between the current block of a first frame and its reference block of a video frame that is temporally related to the first frame. In examples, the 2D pixel location difference may use subpixel interpolation so as to define motion between frames by integer pixels, half pixels, quarter pixels, etc. An illustration of calculating a motion vector is illustrated in
Once a motion vector is determined, the motion vector may be provided to a decoder side within the encoding system. When the decoder receives this information, the decoder may find a corresponding location of a first block on a reference frame that may be linked to a block that is being processed. In this way, the motion vector may be used by the decoder to find a reference. Subsequently, the difference between the reference and the current block (e.g., the motion vector) may be processed and transmitted.
Header information coding may also be used to efficiently encode video data. In particular, header information that is related to a motion vector and header information that is related to a skip mode may be used to encode video data that is captured by a UAV.
Regarding motion vectors, a current block and its spatial neighboring block within the same video frame may have a high probability of sharing the same motion vectors. Moreover, the motion vector temporally corresponding to a current block may also serve as a predictor of the motion vector of the current block. As such, a motion vector predictor (MVP) for a current block may be calculated based on a current block's spatially and temporally neighboring blocks. The calculation of a MVP may depend on the standards of an encoding processor.
Additionally, regarding a skip mode, additional information that is within a header of a current block may also be predicted from neighboring blocks. Further, in examples where a current block may be fully predicted from its neighboring blocks, the header of the current block may be marked as a skip block. In particular, a skip block may be used to indicate that no residual information is transmitted. In examples, a skip may be used when the information within the current block may be calculated based on the information of blocks that neighbor the current block.
As referenced briefly, the difference between an original block and its predictor may be called the residual. In examples, the residual between blocks may be represented as a coefficient. In examples, this residual coefficient may be directly encoded.
In other examples, even after a residual coefficient has been calculated, the residual coefficient may have some spatial redundancy that may be able to be removed. In examples, a transformation may be used to concentrate the energy to low frequency. The concentrated energy may lose less when the video encoding undergoes a processing step which uses lossy compression tools within whatever standard, quantization. For example, if ten numbers [1,1,1,1,1,1,1,1,1,1], are quantized with 10, the ten numbers would become 0, thus causing an error rate of 100%. In contrast, if a transform is applied, the ten numbers would become [10,0,0,0,0,0,0,0,0,0]. If that result is then quantized with 10, nothing is lost, thereby having an error rate that is 0%. This is useful as quantization is a useful lossy compression tool. As such, controlling the quantization parameter may allow control of bitrates of whole streams of data.
While a video encoding process is generally discussed in
Accordingly, when searching for the motion vector that links block 515 to an associated block on the second frame, a block 525 of the second frame that co-located with block 515 of the first frame may be identified. Block 525 may be used to define a search area 530 when searching for a motion vector that links block 515 of the first frame to a block within the second frame. In particular, search area 530 is based on an initial search point 522 at the top left corner of the co-located block, and then extends downwards and to the right of the co-located block so as to cover the area within the co-located block as well as additional area that may also be searched. In examples, the search area may extend downwards and to the right; downwards; downwards and to the left; to the left; upwards and to the right; upwards; upwards and to the left; to the right; or to a combination of these directions relative to an initial starting point. In examples, a starting point may be on a corner of the block or within the block, such as a center of the block. The search area may cover an entirety of the block, cover a portion of the block, overlap with the block, or may at least partially not overlap with the block.
When evaluating the second frame 520 for a motion vector to link a second block with the encoded first block 515, the search area 530 on the second frame 520 may be assessed. In particular, a motion estimation predication may be formed between a first video frame, 510, that has already been coded (also referred to as “coded(510)”) and a second, adjacent frame, 520, so as to form a prediction of 520 (also referred to as “pred(520)”). The adjacent frame, 520, may or may not be encoded. Additionally, the adjacent frame, 520, may be the frame captured immediately after the first video frame, 510, in the video. In some embodiments, the first video frame, 510, may be compared with a subsequent frame, xt, which may or may not be the adjacent frame. The subsequent frame may be the adjacent frame (t=1), or may be the frame immediately after the adjacent frame (t=2), or any frame afterwards (e.g., t=3, 4, 5, 6, 7, 8, 9, 10, or more). In some instances, the subsequent frame may be suitably close in time to the first frame to make a prediction of the next frame “pred(xt).” For instance, t may be 50 or less, 40 or less, 30 or less, 20 or less, 10 or less, 7 or less, 5 or less, 3 or less, 2 or less, or 1. The subsequent frame may be captured within less than 5 seconds of the first frame. In examples, the subsequent frame may be captured within less than 5 seconds, less than 4 seconds, less than 3 seconds, less than 2 seconds, less than 1 second, less than 0.9 seconds, less than 0.8 seconds, less than 0.7 seconds, less than 0.6 seconds, less than 0.5 seconds, less than 0.4 seconds, less than 0.3 seconds, less than 0.2 seconds, less than 0.1 seconds, less than 0.05 seconds, and less than 0.01 seconds.
Any description herein of the adjacent frame, 520, may also apply to a subsequent frame, xt. This motion estimation prediction may be illustrated between the encoded frame 510 and the predicted frame 520 as seen in
For conventional video encoding processes, an assumption is made that the block on first video frame, 510, that correlates with an associated block within the adjacent frame, 520, is located within the same area of the frame. As such, a standard search area may be generated around a block that is co-located with an already encoded block of a temporally adjacent video frame. An example of a standard search area that is generated around a co-located block is illustrated as search area 530 in
However, the search area provided in
Further, the size and shape of the search area may be independent of the motion estimation prediction between frame 510 and frame 520. In conventional motion estimation evaluation processes, once the initial search point, such as search point 522, and search area, such as search area 530, has been determined for frame 520, the motion estimation predication may evaluate the pixels or subpixels of the search area to identify the pixels or subpixels that are most similar to the block within first frame 510. In examples, the similarity of the block on frame 520 to its corresponding block on frame 510 may be estimated by an equation. For example, the similarity of corresponding blocks of frames 510 and 520 may be illustrated by an equation that shows a block of first frame 510 has a corresponding block that is a certain distance from a corresponding block on the adjacent frame 520.
In order to avoid the incorrect association of a block within a first frame and a block within a second, adjacent frame,
As discussed herein,
Accordingly, the encoding processor may adjust the size of a search area associated with a block of frame 620 by analyzing an optical flow field that is generated based on sensor information associated with the UAV. In particular, the optical flow field may be generated based on motion of the camera. Additionally, the optical flow field data may be indicative of an adjustment of a FOV setting of a video capture device, such as a camera. Examples of how an optical flow field may adjust based upon movement of the video capture device and/or movement of an aerial vehicle is provided in
In examples, the encoding processor may increase the size of the search area by 0.1%, 0.2%, 0.3%, 0.4%, 0.5%, 0.60, 0.7%, 0.8%, 0.9%, 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 100%, 120%, 140%, 160%, 180%, 200%, 250%, 500%, or more than 500%. Alternatively, the encoding processor may decrease the size of the search area by 0.1%, 0.2%, 0.3%, 0.4%, 0.5%, 0.6%, 0.7%, 0.8%, 0.9%, 1%, 2%, 3%, 4%, 5%, 6%, 7%, 8%, 9%, 10%, 15%, 20%, 25%, 30%, 35%, 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 95%, 100%, 120%, 140%, 160%, 180%, 200%, 250%, 500%, or more than 500%.
Accordingly, the encoder may adjust the shape of a search area based on optical flow fields that are generated based on sensor data associated with the UAV. In examples, the search area may be a shape such as a circle, square, rectangle, triangle, rhombus, parallelogram, trapezoid, oval, ellipse, figure-8, or another shape.
The encoder may also move the search area to a new location. In particular, the encoder may move the search area from its initial starting position in a given direction. In examples, the encoder may move the search area to the left, to the right, up, down, or at an angle from the starting position of the search area.
Additionally, the encoder may adjust the dimensions of a search area based on optical flow fields that are generated based sensor data associated with the UAV. An optical flow field may describe the direction and/or velocity of points on a frame. In some examples, an optical flow field may describe the direction and velocity of all points on a frame. Sensors that are associated with the UAV may be used to generate an optical flow field. In particular, sensors that may be used for generating an optical flow field may include a GPS, a gyroscope, a MVO, the angle of the camera, height information, or any other sensors described herein.
While an increase in search area size, as seen in
Optical flow fields that are generated may be used by an encoding process to increase the accuracy of the initial motion vector. Accordingly, the size, shape, and/or location of a search area may be modified based on an optical flow field that is generated based on sensor data associated with the UAV.
An example of an optical flow field 700 is provided in
Additionally, the distance that the modified search area 720 is translated may be based on how fast the object seems to be moving relative to the video captured on the image capture device. This speed may be assessed based on the distance that is travelled of similar objects as contextualized by time stamps that are associated with the video frames. In this way, distance of object movement in context of timestamps (so, e.g., elapsed time) may indicate speed. Additionally, as provided in
The relationship of the perceived size of objects within an optical flow field may vary based on location of the objects within the optical flow field. For example, when an optical flow field is generated based on a zoom-in action, objects that are the same size in real life may appear to be larger as they are located further to the edge of the optical flow field. This is illustrated in
An example of an optical flow field that has different rates of movement associated with objects within a frame is provided in
In contrast, some objects that are in the central and lower part of the optical flow field are moving relatively fast. In particular, objects may seem to move relatively fast based on their movement relative to a video capture device. In particular, if a video capture device is moving quickly past a stationary object, the stationary object may seem to be moving quickly based on the movement of the video capture device. In examples, the perceived movement of objects may have a motion component that is associated with movement of the video capture device and/or may have a motion component that is associated with movement of a movable object, such as UAV, that attaches the video capture device. Additionally,
Additionally,
As such, an initial search area 1110 may be translated along motion areas within optical flow field 1100 so as to illustrate modified search area 1120. The degree of translation may be based on the speed of objects between frames. For example, if objects within the video are moving at a relatively slow speed, the modified search area may be located relatively close to the initial search area. However, if objects within the video are moving at a relatively fast speed, the modified search area may be located relatively far from the initial search area. As seen in
Additional optical flow fields may be generated by rotating a video capture device about a roll axis, by rotating a UAV about camera roll axis, or both. In additional examples, generated optical flow fields may include motion aspects related to the speed of moving objects, distance of moving objects from a video capture device, curving motion of moving objects, directionality of moving objects, and other characteristics of object movement within an optical flow field.
In another example, an illustration 1200 of a bias of a search area is provided in
As seen in
In examples, motion within an optical flow field may be assessed using a vector named MVo=[XMVo, YMVo]. Additionally, the accuracy of the motion vector may be determined to be γ. To make use of this prior knowledge, the initial point may be biased by the amount of γ*Mo.
Additionally, for a search range, a set of searching ranges may be predefined. In examples, the shape of a search range may not be restricted to square. However, a square shape may be used in examples. As the search range is of two dimensions, each dimension may have three scales. In examples for a square search range, a width direction may have the dimensions Wsmall, Wnormal and Wlarge. Additionally, a vertical direction may have the dimensions Hsmall, Hnormal and Hlarge. In this way, a 2-dimensional search range may have 9 types of search windows. In examples, a 2-dimensional search range may have a number of pre-defined searching sizes, such as 50 or fewer, 40 or fewer, 30 or fewer, 25 or fewer, 20 or fewer, 15 or fewer, 14 or fewer, 13 or fewer, 12 or fewer, 10 or fewer, 9 or fewer, 8 or fewer, 7 or fewer, 6 or fewer, 5 or fewer, 4 or fewer, 3 or fewer, 2 or fewer, 1 pre-defined search seize, or the 2-dimensional search range may not have a pre-defined search size.
Additionally, a predefined threshold may be used to determine whether the motion is violent. In examples, thresholds may be denoted as θlow and θhigh. Additionally, the search windows dimension [W, H] may be determined using the relationship related to width:
Additionally, a calculation of a height H may be conducted in a similar manner. In examples, when the initial search point is biased and the search window is enlarged, there may be a higher possibility to find the best motion vector.
In the above case, the motion on the whole frame may be unique. However, in some other cases, the initial motion vector and search range may adapt to the optical flow field globally and also locally. In some cases, the motion on the top part of the video frame may be more still than the bottom part of the same video frame. In some other cases, objects that are far away from a frame center may move faster than objects that are closer to the frame center.
At block 1620, motion data that is associated with the second video frame is received. The motion data are captured by sensors. In particular, the sensors may be within the image capture device. Alternatively, the sensors may be remote from the image capture device. At block 1630, a search area within the first video frame is determined based on the motion data that is associated with the second video frame. In examples, the size of the search area may be determined based on the motion data that is associated with the second video frame. In further examples, the location of the search area on the first video frame may be determined based on the motion data that is associated with the second video frame. Additionally, the search area is used to evaluate motion of the video captured by the image capture device. In examples, the search area is used to evaluate motion of the video captured by the image capture device to identify a motion vector within the video.
The systems, devices, and methods described herein for video encoding may apply to any video that is captured by a video capture device supported by a variety of objects. In particular, the video may be captured by a video capture device that is supported by an aerial vehicle. As previously mentioned, any description herein of an aerial vehicle, such as a UAV, may apply to and be used for any movable object. Any description herein of an aerial vehicle may apply specifically to UAVs. A movable object of the present disclosure may be configured to move within any suitable environment, such as in air (e.g., a fixed-wing aircraft, a rotary-wing aircraft, or an aircraft having neither fixed wings nor rotary wings), in water (e.g., a ship or a submarine), on ground (e.g., a motor vehicle, such as a car, truck, bus, van, motorcycle, bicycle; a movable structure or frame such as a stick, fishing pole; or a train), under the ground (e.g., a subway), in space (e.g., a spaceplane, a satellite, or a probe), or any combination of these environments. The movable object may be a vehicle, such as a vehicle described elsewhere herein. In some embodiments, the movable object may be carried by a living subject, or take off from a living subject, such as a human or an animal. Suitable animals may include avines, canines, felines, equines, bovines, ovines, porcines, delphines, rodents, or insects.
The movable object may be capable of moving freely within the environment with respect to six degrees of freedom (e.g., three degrees of freedom in translation and three degrees of freedom in rotation). Alternatively, the movement of the movable object may be constrained with respect to one or more degrees of freedom, such as by a predetermined path, track, or orientation. The movement may be actuated by any suitable actuation mechanism, such as an engine or a motor. The actuation mechanism of the movable object may be powered by any suitable energy source, such as electrical energy, magnetic energy, solar energy, wind energy, gravitational energy, chemical energy, nuclear energy, or any suitable combination thereof. The movable object may be self-propelled via a propulsion system, as described elsewhere herein. The propulsion system may optionally run on an energy source, such as electrical energy, magnetic energy, solar energy, wind energy, gravitational energy, chemical energy, nuclear energy, or any suitable combination thereof. Alternatively, the movable object may be carried by a living being.
In some instances, the movable object may be an aerial vehicle. For example, aerial vehicles may be fixed-wing aircraft (e.g., airplane, gliders), rotary-wing aircraft (e.g., helicopters, rotorcraft), aircraft having both fixed wings and rotary wings, or aircraft having neither (e.g., blimps, hot air balloons). An aerial vehicle may be self-propelled, such as self-propelled through the air. A self-propelled aerial vehicle may utilize a propulsion system, such as a propulsion system including one or more engines, motors, wheels, axles, magnets, rotors, propellers, blades, nozzles, or any suitable combination thereof. In some instances, the propulsion system may be used to enable the movable object to take off from a surface, land on a surface, maintain its current position and/or orientation (e.g., hover), change orientation, and/or change position.
The movable object may be controlled remotely by a user or controlled locally by an occupant within or on the movable object. The movable object may be controlled remotely via an occupant within a separate vehicle. In some embodiments, the movable object is an unmanned movable object, such as a UAV. An unmanned movable object, such as a UAV, may not have an occupant on-board the movable object. The movable object may be controlled by a human or an autonomous control system (e.g., a computer control system), or any suitable combination thereof. The movable object may be an autonomous or semi-autonomous robot, such as a robot configured with an artificial intelligence.
The movable object may have any suitable size and/or dimensions. In some embodiments, the movable object may be of a size and/or dimensions to have a human occupant within or on the vehicle. Alternatively, the movable object may be of size and/or dimensions smaller than that capable of having a human occupant within or on the vehicle. The movable object may be of a size and/or dimensions suitable for being lifted or carried by a human. Alternatively, the movable object may be larger than a size and/or dimensions suitable for being lifted or carried by a human. In some instances, the movable object may have a maximum dimension (e.g., length, width, height, diameter, diagonal) of less than or equal to about: 2 cm, 5 cm, 10 cm, 50 cm, 1 m, 2 m, 5 m, or 10 m. The maximum dimension may be greater than or equal to about: 2 cm, 5 cm, 10 cm, 50 cm, 1 m, 2 m, 5 m, or 10 m. For example, the distance between shafts of opposite rotors of the movable object may be less than or equal to about: 2 cm, 5 cm, 10 cm, 50 cm, 1 m, 2 m, 5 m, or 10 m. Alternatively, the distance between shafts of opposite rotors may be greater than or equal to about: 2 cm, 5 cm, 10 cm, 50 cm, 1 m, 2 m, 5 m, or 10 m.
In some embodiments, the movable object may have a volume of less than 100 cm×100 cm×100 cm, less than 50 cm×50 cm×30 cm, or less than 5 cm×5 cm×3 cm. The total volume of the movable object may be less than or equal to about: 1 cm3, 2 cm3, 5 cm3, 10 cm3, 20 cm3, 30 cm3, 40 cm3, 50 cm3, 60 cm3, 70 cm3, 80 cm3, 90 cm3, 100 cm3, 150 cm3, 200 cm3, 300 cm3, 500 cm3, 750 cm3, 1000 cm3, 5000 cm3, 10,000 cm3, 100,000 cm33, 1 m3, or 10 m3. Conversely, the total volume of the movable object may be greater than or equal to about: 1 cm3, 2 cm3, 5 cm3, 10 cm3, 20 cm3, 30 cm3, 40 cm3, 50 cm3, 60 cm3, 70 cm3, 80 cm3, 90 cm3, 100 cm3, 150 cm3, 200 cm3, 300 cm3, 500 cm3, 750 cm3, 1000 cm3, 5000 cm3, 10,000 cm3, 100,000 cm3, 1 m3, or 10 m3.
In some embodiments, the movable object may have a footprint (which may refer to the lateral cross-sectional area encompassed by the movable object) less than or equal to about: 32,000 cm2, 20,000 cm2, 10,000 cm2, 1,000 cm2, 500 cm2, 100 cm2, 50 cm2, 10 cm2, or 5 cm2. Conversely, the footprint may be greater than or equal to about: 32,000 cm2, 20,000 cm2, 10,000 cm2, 1,000 cm2, 500 cm2, 100 cm2, 50 cm2, 10 cm2, or 5 cm2.
In some instances, the movable object may weigh no more than 1000 kg. The weight of the movable object may be less than or equal to about: 1000 kg, 750 kg, 500 kg, 200 kg, 150 kg, 100 kg, 80 kg, 70 kg, 60 kg, 50 kg, 45 kg, 40 kg, 35 kg, 30 kg, 25 kg, 20 kg, 15 kg, 12 kg, 10 kg, 9 kg, 8 kg, 7 kg, 6 kg, 5 kg, 4 kg, 3 kg, 2 kg, 1 kg, 0.5 kg, 0.1 kg, 0.05 kg, or 0.01 kg. Conversely, the weight may be greater than or equal to about: 1000 kg, 750 kg, 500 kg, 200 kg, 150 kg, 100 kg, 80 kg, 70 kg, 60 kg, 50 kg, 45 kg, 40 kg, 35 kg, 30 kg, 25 kg, 20 kg, 15 kg, 12 kg, 10 kg, 9 kg, 8 kg, 7 kg, 6 kg, 5 kg, 4 kg, 3 kg, 2 kg, 1 kg, 0.5 kg, 0.1 kg, 0.05 kg, or 0.01 kg.
In some embodiments, a movable object may be small relative to a load carried by the movable object. The load may include a payload and/or a carrier, as described in further detail elsewhere herein. In some examples, a ratio of a movable object weight to a load weight may be greater than, less than, or equal to about 1:1. In some instances, a ratio of a movable object weight to a load weight may be greater than, less than, or equal to about 1:1. Optionally, a ratio of a carrier weight to a load weight may be greater than, less than, or equal to about 1:1. When desired, the ratio of an movable object weight to a load weight may be less than or equal to: 1:2, 1:3, 1:4, 1:5, 1:10, or even less. Conversely, the ratio of a movable object weight to a load weight may also be greater than or equal to: 2:1, 3:1, 4:1, 5:1, 10:1, or even greater.
In some embodiments, the movable object may have low energy consumption. For example, the movable object may use less than about: 5 W/h, 4 W/h, 3 W/h, 2 W/h, 1 W/h, or less. In some instances, a carrier of the movable object may have low energy consumption. For example, the carrier may use less than about: 5 W/h, 4 W/h, 3 W/h, 2 W/h, 1 W/h, or less. Optionally, a payload of the movable object may have low energy consumption, such as less than about: 5 W/h, 4 W/h, 3 W/h, 2 W/h, 1 W/h, or less.
In some embodiments, the movable object may be configured to carry a load. The load may include one or more of passengers, cargo, equipment, instruments, and the like. The load may be provided within a housing. The housing may be separate from a housing of the movable object, or be part of a housing for a movable object. Alternatively, the load may be provided with a housing while the movable object does not have a housing. Alternatively, portions of the load or the entire load may be provided without a housing. The load may be rigidly fixed relative to the movable object. Optionally, the load may be movable relative to the movable object (e.g., translatable or rotatable relative to the movable object). The load may include a payload and/or a carrier, as described elsewhere herein.
In some embodiments, the movement of the movable object, carrier, and payload relative to a fixed reference frame (e.g., the surrounding environment) and/or to each other, may be controlled by a terminal. The terminal may be a remote control device at a location distant from the movable object, carrier, and/or payload. The terminal may be disposed on or affixed to a support platform. Alternatively, the terminal may be a handheld or wearable device. For example, the terminal may include a smartphone, tablet, laptop, computer, glasses, gloves, helmet, microphone, or suitable combinations thereof. The terminal may include a user interface, such as a keyboard, mouse, joystick, touchscreen, or display. Any suitable user input may be used to interact with the terminal, such as manually entered commands, voice control, gesture control, or position control (e.g., via a movement, location or tilt of the terminal).
The terminal may be used to control any suitable state of the movable object, carrier, and/or payload. For example, the terminal may be used to control the position and/or orientation of the movable object, carrier, and/or payload relative to a fixed reference from and/or to each other. In some embodiments, the terminal may be used to control individual elements of the movable object, carrier, and/or payload, such as the actuation assembly of the carrier, a sensor of the payload, or an emitter of the payload. The terminal may include a wireless communication device adapted to communicate with one or more of the movable object, carrier, or payload.
The terminal may include a suitable display unit for viewing information of the movable object, carrier, and/or payload. For example, the terminal may be configured to display information of the movable object, carrier, and/or payload with respect to position, translational velocity, translational acceleration, orientation, angular velocity, angular acceleration, or any suitable combinations thereof. In some embodiments, the terminal may display information provided by the payload, such as data provided by a functional payload (e.g., images recorded by a camera or other image capturing device).
Optionally, the same terminal may both control the movable object, carrier, and/or payload, or a state of the movable object, carrier and/or payload, as well as receive and/or display information from the movable object, carrier and/or payload. For example, a terminal may control the positioning of the payload relative to an environment, while displaying image data captured by the payload, or information about the position of the payload. Alternatively, different terminals may be used for different functions. For example, a first terminal may control movement or a state of the movable object, carrier, and/or payload while a second terminal may receive and/or display information from the movable object, carrier, and/or payload. For example, a first terminal may be used to control the positioning of the payload relative to an environment while a second terminal displays image data captured by the payload. Various communication modes may be utilized between a movable object and an integrated terminal that both controls the movable object and receives data, or between the movable object and multiple terminals that both control the movable object and receives data. For example, at least two different communication modes may be formed between the movable object and the terminal that both controls the movable object and receives data from the movable object.
The propulsion mechanisms 1906 may include one or more of rotors, propellers, blades, engines, motors, wheels, axles, magnets, or nozzles, as previously described. The movable object may have one or more, two or more, three or more, or four or more propulsion mechanisms. The propulsion mechanisms may all be of the same type. Alternatively, one or more propulsion mechanisms may be different types of propulsion mechanisms. The propulsion mechanisms 1906 may be mounted on the movable object 1900 using any suitable means, such as a support element (e.g., a drive shaft) as described elsewhere herein. The propulsion mechanisms 1906 may be mounted on any suitable portion of the movable object 1900, such on the top, bottom, front, back, sides, or suitable combinations thereof.
In some embodiments, the propulsion mechanisms 1906 may enable the movable object 1900 to take off vertically from a surface or land vertically on a surface without requiring any horizontal movement of the movable object 1900 (e.g., without traveling down a runway). Optionally, the propulsion mechanisms 1906 may be operable to permit the movable object 1900 to hover in the air at a specified position and/or orientation. One or more of the propulsion mechanisms 1900 may be controlled independently of the other propulsion mechanisms.
Alternatively, the propulsion mechanisms 1900 may be configured to be controlled simultaneously. For example, the movable object 1900 may have multiple horizontally oriented rotors that may provide lift and/or thrust to the movable object. The multiple horizontally oriented rotors may be actuated to provide vertical takeoff, vertical landing, and hovering capabilities to the movable object 1900. In some embodiments, one or more of the horizontally oriented rotors may spin in a clockwise direction, while one or more of the horizontally rotors may spin in a counterclockwise direction. For example, the number of clockwise rotors may be equal to the number of counterclockwise rotors. The rotation rate of each of the horizontally oriented rotors may be varied independently in order to control the lift and/or thrust produced by each rotor, and thereby adjust the spatial disposition, velocity, and/or acceleration of the movable object 1900 (e.g., with respect to up to three degrees of translation and up to three degrees of rotation).
The sensing system 1908 may include one or more sensors that may sense the spatial disposition, velocity, and/or acceleration of the movable object 1900 (e.g., with respect to up to three degrees of translation and up to three degrees of rotation). The one or more sensors may include global positioning system (GPS) sensors, motion sensors, inertial sensors, proximity sensors, or image sensors. The sensing data provided by the sensing system 1908 may be used to control the spatial disposition, velocity, and/or orientation of the movable object 1900 (e.g., using a suitable processing unit and/or control module, as described below). Alternatively, the sensing system 1908 may be used to provide data regarding the environment surrounding the movable object, such as weather conditions, proximity to potential obstacles, location of geographical features, location of manmade structures, and the like.
The communication system 1910 enables communication with terminal 1912 having a communication system 1914 via wireless signals 1916. The communication systems 1910, 1914 may include any number of transmitters, receivers, and/or transceivers suitable for wireless communication. The communication may be one-way communication, such that data may be transmitted in only one direction. For example, one-way communication may involve only the movable object 1900 transmitting data to the terminal 1912, or vice-versa. The data may be transmitted from one or more transmitters of the communication system 1910 to one or more receivers of the communication system 1912, or vice-versa. Alternatively, the communication may be two-way communication, such that data may be transmitted in both directions between the movable object 1900 and the terminal 1912. The two-way communication may involve transmitting data from one or more transmitters of the communication system 1910 to one or more receivers of the communication system 1914, and vice-versa.
In some embodiments, the terminal 1912 may provide control data to one or more of the movable object 1900, carrier 1902, and payload 1904 and receive information from one or more of the movable object 1900, carrier 1902, and payload 1904 (e.g., position and/or motion information of the movable object, carrier or payload; data sensed by the payload such as image data captured by a payload camera). In some instances, control data from the terminal may include instructions for relative positions, movements, actuations, or controls of the movable object, carrier and/or payload. For example, the control data may result in a modification of the location and/or orientation of the movable object (e.g., via control of the propulsion mechanisms 1906), or a movement of the payload with respect to the movable object (e.g., via control of the carrier 1902). The control data from the terminal may result in control of the payload, such as control of the operation of a camera or other image capturing device (e.g., taking still or moving pictures, zooming in or out, turning on or off, switching imaging modes, change image resolution, changing focus, changing depth of field, changing exposure time, changing viewing angle or field of view). In some instances, the communications from the movable object, carrier and/or payload may include information from one or more sensors (e.g., of the sensing system 1908 or of the payload 1904). The communications may include sensed information from one or more different types of sensors (e.g., GPS sensors, motion sensors, inertial sensor, proximity sensors, or image sensors). Such information may pertain to the position (e.g., location, orientation), movement, or acceleration of the movable object, carrier and/or payload. Such information from a payload may include data captured by the payload or a sensed state of the payload. The control data provided transmitted by the terminal 1912 may be configured to control a state of one or more of the movable object 1900, carrier 1902, or payload 1904. Alternatively or in combination, the carrier 1902 and payload 1904 may also each include a communication module configured to communicate with terminal 1912, such that the terminal may communicate with and control each of the movable object 1900, carrier 1902, and payload 1904 independently.
In some embodiments, the movable object 1900 may be configured to communicate with another remote device in addition to the terminal 1912, or instead of the terminal 1912. The terminal 1912 may also be configured to communicate with another remote device as well as the movable object 1900. For example, the movable object 1900 and/or terminal 1912 may communicate with another movable object, or a carrier or payload of another movable object. When desired, the remote device may be a second terminal or other computing device (e.g., computer, laptop, tablet, smartphone, or other mobile device). The remote device may be configured to transmit data to the movable object 1900, receive data from the movable object 1900, transmit data to the terminal 1912, and/or receive data from the terminal 1912. Optionally, the remote device may be connected to the Internet or other telecommunications network, such that data received from the movable object 1900 and/or terminal 1912 may be uploaded to a website or server.
The sensing module 2002 may utilize different types of sensors that collect information relating to the movable objects in different ways. Different types of sensors may sense different types of signals or signals from different sources. For example, the sensors may include inertial sensors, GPS sensors, proximity sensors (e.g., lidar), or vision/image sensors (e.g., a camera). The sensing module 2002 may be operatively coupled to a processing unit 2004 having a plurality of processors. In some embodiments, the sensing module may be operatively coupled to a transmission module 2012 (e.g., a Wi-Fi image transmission module) configured to directly transmit sensing data to a suitable external device or system. For example, the transmission module 2012 may be used to transmit images captured by a camera of the sensing module 2002 to a remote terminal.
The processing unit 2004 may have one or more processors, such as a programmable processor (e.g., a central processing unit (CPU)). The processing unit 2004 may be operatively coupled to a non-transitory computer readable medium 2006. The non-transitory computer readable medium 2006 may store logic, code, and/or program instructions executable by the processing unit 2004 for performing one or more steps. The non-transitory computer readable medium may include one or more memory units (e.g., removable media or external storage such as an SD card or random access memory (RAM)). In some embodiments, data from the sensing module 2002 may be directly conveyed to and stored within the memory units of the non-transitory computer readable medium 2006. The memory units of the non-transitory computer readable medium 2006 may store logic, code and/or program instructions executable by the processing unit 2004 to perform any suitable embodiment of the methods described herein. For example, the processing unit 2004 may be configured to execute instructions causing one or more processors of the processing unit 2004 to analyze sensing data produced by the sensing module. The memory units may store sensing data from the sensing module to be processed by the processing unit 2004. In some embodiments, the memory units of the non-transitory computer readable medium 2006 may be used to store the processing results produced by the processing unit 2004.
In some embodiments, the processing unit 2004 may be operatively coupled to a control module 2008 configured to control a state of the movable object. For example, the control module 2008 may be configured to control the propulsion mechanisms of the movable object to adjust the spatial disposition, velocity, and/or acceleration of the movable object with respect to six degrees of freedom. Alternatively or in combination, the control module 2008 may control one or more of a state of a carrier, payload, or sensing module.
The processing unit 2004 may be operatively coupled to a communication module 2010 configured to transmit and/or receive data from one or more external devices (e.g., a terminal, display device, or other remote controller). Any suitable means of communication may be used, such as wired communication or wireless communication. For example, the communication module 2010 may utilize one or more of local area networks (LAN), wide area networks (WAN), infrared, radio, WiFi, point-to-point (P2P) networks, telecommunication networks, cloud communication, and the like. Optionally, relay stations, such as towers, satellites, or mobile stations, may be used. Wireless communications may be proximity dependent or proximity independent. In some embodiments, line-of-sight may or may not be required for communications. The communication module 2010 may transmit and/or receive one or more of sensing data from the sensing module 2002, processing results produced by the processing unit 2004, predetermined control data, user commands from a terminal or remote controller, and the like.
The components of the system 2000 may be arranged in any suitable configuration. For example, one or more of the components of the system 2000 may be located on the movable object, carrier, payload, terminal, sensing system, or an additional external device in communication with one or more of the above. Additionally, although
While some embodiments of the present disclosure have been shown and described herein, it will be obvious to those skilled in the art that such embodiments are provided by way of example only. Numerous variations, changes, and substitutions will now occur to those skilled in the art without departing from the invention. It should be understood that various alternatives to the embodiments of the disclosure described herein may be employed in practicing the disclosure. It is intended that the following claims define the scope of the invention and that methods and structures within the scope of these claims and their equivalents be covered thereby.
Claims
1. A video encoding method comprising:
- receiving a video captured by an image capture device on-board a movable object, the video including a video frame component;
- receiving sensor data from a plurality of sensors on-board the movable object;
- generating, according to the sensor data, an optical flow field associated with the video frame component; and
- evaluating motion of the video for video encoding based on the optical flow field.
2. The method of claim 1, wherein evaluating the motion of the video includes:
- identifying an object within the video frame component; and
- assessing, based on the optical flow filed, a motion vector of the object.
3. The method of claim 2, further comprising determining an initial point for assessing the motion vector within the video frame component.
4. The method of claim 3, wherein:
- the video frame component is a first video frame component of the video; and
- determining the initial point includes determining the initial point by referencing a second video frame component of the video adjacent to the first video frame component.
5. The method of claim 1, wherein evaluating the motion of the video includes:
- determining, based on the optical flow field, a portion of the motion of the video that is caused by a movement of the image capture device.
6. The method of claim 1, wherein the plurality of sensors include at least one of an image sensor, a GPS, an inertial measurement unit, a lidar, an ultrasonic sensor, an acoustic sensor, or a WiFi sensor.
7. The method of claim 1, wherein the sensor data includes information about a spatial disposition of the image capture device, a movement of the image capture device, a zoom state of the image capture device, and a field of view (FOV) of the image capture device.
8. The method of claim 1, wherein generating the optical flow field includes generating the optical flow filed further according to data received from a source external to the movable object.
9. The method of claim 8, wherein the data received from the source external to the movable object include data transmitted from a ground station to the movable object.
10. The method of claim 1, wherein evaluating the motion of the video includes evaluating the motion of the video within a search area in the video frame component of the video.
11. The method of claim 10, further comprising:
- determining a dimension, shape, or location of the search area based on the optical flow field.
12. The method of claim 11, further comprising:
- obtaining an initial search area within the video frame component;
- determining, based on the optical flow filed, a bias of the initial search area;
- adjusting, based on the optical flow field, the initial search area to determine the search area within the video frame component.
13. The method of claim 12, wherein adjusting the initial search area includes vertically elongating the initial search area when the optical flow field indicates a vertical bias.
14. The method of claim 12, wherein adjusting the initial search area includes moving a position of the initial search area on the video frame component along a circular arc when the optical flow field indicates a rotation.
15. The method of claim 12,
- wherein the video includes one or more objects;
- the method further comprising: determining, based on the optical flow field, whether the image capture device moves with respect to the one or more objects faster in the one direction than in another direction; and in response to determining the image capture device moves with respect to the one or more objects faster in the one direction than in the another direction, adjusting the boundary of the initial search area by a first ratio in the one direction and adjusting the boundary of the initial search area by a second ratio in the another direction, the first ratio being greater than the second ratio.
16. The method of claim 12, wherein a dimension of the search area is smaller than a dimension of the initial search area.
17. The method of claim 1, wherein the movable object is an unmanned aerial vehicle.
18. A video encoding system comprising:
- an image capture device on-board a movable object and configured to capture a video;
- one or more sensors on-board the movable object and configured to obtain sensor data; and
- one or more processors, individually or collectively configured to: receive the video from the image capture device, the video including a video frame component; receive the sensor data from the one or more sensors; generate, according to the sensor data, an optical flow field associated with the video frame component; evaluate motion of the video for video encoding based on the optical flow field.
19. The system of claim 18, further comprising a gimbal configured to couple the image capture device to the movable object.
20. The system of claim 19, wherein at least a portion of the plurality of sensors is arranged at the gimbal.
Type: Application
Filed: Jun 25, 2020
Publication Date: Oct 15, 2020
Inventors: Yannan WU (Shenzhen), Xiaozheng TANG (Shenzhen), Wei CHEN (Shenzhen), Zisheng CAO (Shenzhen), Mingyu WANG (Shenzhen)
Application Number: 16/912,044
