Abstract: Systems and methods in accordance with embodiments of the invention estimate depth from projected texture using camera arrays that includes at least two two-dimensional arrays of cameras each several cameras; an illumination system configured to illuminate a scene with a projected texture; a processor; and memory containing an image processing pipeline application and an illumination system controller application. In addition, the illumination system controller application directs the processor to control the illumination system to illuminate a scene with a projected texture. Furthermore, the image processing pipeline application directs the processor to: utilize the illumination system controller application to control the illumination system to illuminate a scene with a projected texture capture a set of images of the scene illuminated with the projected texture; determining depth estimates for pixel locations in an image from a reference viewpoint using at least a subset of the set of images.

Abstract: This invention provides a single-camera vision system, typically for use in logistics applications, that allows for adjustment of the camera viewing angle to accommodate a wide range of object heights and associated widths moving relative to an imaged scene with constant magnification. The camera assembly employs an image sensor that is more particularly suited to such applications, with an aspect (height-to-width) ratio of approximately 1:4 to 1:8. The camera assembly includes a distance sensor to determine the distance to the top of each object. The camera assembly employs a zoom lens that can change at relatively high speed (e.g. <10 ms) to allow adjustment of the viewing angle from object to object as each one passes under the camera's field of view (FOV). Optics that allow the image to be resolved on the image sensor within the desired range of viewing angles are provided in the camera lens assembly.

Abstract: Presented herein are techniques for filtering pixels during video coding and decoding operations. Similar operations are performed at a video encoder and a video decoder. For a target pixel in a block of a video frame represented by the encoded bit-stream, a value of the target pixel is compared with neighboring pixels to produce a plurality of comparison results. A particular offset value for the target pixel is derived based on the plurality of comparison results. The target pixel is filtered using the particular offset value. This process is performed for some or all of the pixels of blocks of a video frame.

Abstract: Systems and methods for determining a target number of bits (target bitrate) for encoding a frame of video that will satisfy a buffer constraint in a parallel video encoder. The quantization parameter (QP) for a given encoding process may be determined for the frame based on the target bitrate to maintain a suitable average bitrate. In some embodiments, the bitrate used for one or more prior frame is estimated. In some embodiments, a buffer fullness update is made based on an estimated bitrate. In some embodiments, a bitrate to target for each frame is determined based on the frame type, estimated bitrate of a prior frame(s), and the updated buffer fullness.

Abstract: Alert directives and focused alert directives allow a user to provide feedback to a behavioral recognition system to always or never publish an alert for certain events. Such an approach bypasses the normal publication methods of the behavioral recognition system yet does not obstruct the system's learning procedures.

Abstract: Device and method for scanning an object outline image are provided. The scanning device includes a light source, an optical sensor and a processor. The scanning method includes: providing a polarized projection light beam; projecting the polarized projection light beam to an object, and correspondingly reflecting a polarized reflection light beam by the object according to the polarized projection light beam; capturing an image of the polarized reflection light beam; calculating polarization state of target according to the polarized projection light beam and the polarized reflection light beam, the polarization state of target having a plane angle from normal projection and a corresponding point position; and restoring an outline image of the object according to the polarization state of target.

Abstract: Variations of rho-domain rate control for video encoding or other media encoding are presented. For example, in some of the variations, an encoder sets a rho value for a unit of media based at least in part on a bit allocation for the unit. The encoder also computes transform coefficients for the unit using a frequency transform having multiple location-dependent scale factors, sets a value of quantization parameter (“QP”) for the unit using a mapping of QP values to rho values, and uses the value of QP for the unit during quantization of the transform coefficients of the unit. When the QP-rho mapping is determined, a location-independent scale factor that approximates the multiple location-dependent scale factors is used and/or certain scaling operations are integrated, which reduces computational complexity while still supporting accurate rate control decisions. Implementations of such variations of rate control can exploit opportunities for caching and parallel computation.

Abstract: A video capture device may include multiple cameras that simultaneously capture video data. The video capture device and/or one or more remote computing resources may stitch the video data captured by the multiple cameras to generate stitched video data that corresponds to 360° video. The remote computing resources may apply one or more algorithms to the stitched video data to adjust the color characteristics of the stitched video data, such as lighting, exposure, white balance contrast, and saturation. The remote computing resources may further smooth the transition between the video data captured by the multiple cameras to reduce artifacts such as abrupt changes in color as a result of the individual cameras of the video capture device having different video capture settings. The video capture device and/or the remote computing resources may generate a panoramic video that may include up to a 360° field of view.

Abstract: Systems, apparatus and methods are described including operations for performing, via a frame division module, frame division of video frames into sections to form a first frame section stream and a second frame section stream; and/or encoding, via a first and a second encoder, the first frame section stream via the first encoder and the second frame section stream via the second encoder.

Abstract: A method of statistical multiplexing digital video is provided enabling use of an encoder pool. The method includes loading a mapping curve into memory at a statistical multiplexor controller (SMC) for each of a plurality of encoders in a heterogeneous encoder pool, each mapping curve being specific to a type and/or configuration of encoder and associating target bitrates with need parameter values, receiving a need parameter value at the SMC from each encoder in the heterogeneous encoder pool, determining a target bitrate associated with each need parameter value received at the SMC, by looking up target bitrates for each need parameter on mapping curves specific to each encoder in the heterogeneous encoder pool, and sending the target bitrate from the SMC to each encoder in the heterogeneous encoder pool in a bitrate assignment.

Abstract: The present invention comprises a foveated imaging system capable of capturing a wide field of view image and a foveated image, where the foveated image is a controllable region of interest of the wide field of view image.

Abstract: Disclosed an apparatus for processing images in a stereo vision system including: an image capturing unit including first and second cameras for obtaining image information by capturing a subject, and a lighting unit for emitting a flash during the image capturing; and a camera controller for controlling the image information to be obtained by cross-acquiring images of the subject in a predetermined time interval without the synchronization of the first and second cameras.

Abstract: The technology disclosed relates to coordinating motion-capture of a hand by a network of motion-capture sensors having overlapping fields of view. In particular, it relates to designating a first sensor among three or more motion-capture sensors as having a master frame of reference, observing motion of a hand as it passes through overlapping fields of view of the respective motion-capture sensors, synchronizing capture of images of the hand within the overlapping fields of view by pairs of the motion-capture devices, and using the pairs of the hand images captured by the synchronized motion-capture devices to automatically calibrate the motion-capture sensors to the master frame of reference frame.

Abstract: A video surveillance system, device and methods may accurately model the shape of a human object monitored by a video stream. 3D human models, such as a coarse 3D human model and a detailed 3D human model may be estimated by mapping individual body part components to a frame. For example, a coarse 3D human model may be obtained by mapping the cylindrical body parts to a plurality of skeleton pose estimates on a part by part basis. A detailed 3D human model may be estimated by mapping detailed human body parts to respective the cylindrical body parts of the coarse 3D human model on a part by part basis. The detailed 3D human model may be used to detect accessories of the human object being monitored, as well as overall dimensions, body part dimensions, age, and gender of the human object being monitored.

Abstract: A system and method for facilitating inhibiting banding in video data, in part by anticipating when banding may occur given certain encoder parameters. An example method includes receiving an input stream of video data; extracting feature information characterizing the stream of video data; using a feed-forward neural network to process the feature information to estimate when a particular block of the stream of video data will exhibit an artifact when encoded using certain Quantization Parameters (QPs); incorporating the indication into metadata associated with the stream of video data; and transferring the stream of video data and metadata to a video encoder. In a more specific embodiment, the feature information includes color information and texture information, and the neural network includes a feed-forward neural network that includes a classifier with a sigmoid activation function, and which has been trained using a cross-entropy cost function.

Abstract: According to an embodiment, A test system includes: a moving unit configured to move a test object, the test object including a first surface, a mark being printed on the first surface; a first imaging device configured to photograph the first surface of test object to obtain a first image; a cutter configured to scratch the first surface; a first unit configured to attach a tape to the first surface; a second unit configured to detach the tape from the first surface; a second imaging device configured to photograph the first surface after detaching the tape to obtain a second image; and a controller configured to compare the first image and the second image to output a comparison result.

Abstract: Embodiments described herein may help to provide methods for sharing and viewing part of an environment of a computing device, such as a head-mountable device (HMD). An example method involves: (a) determining a still photo panorama of an environment, (b) receiving a video stream of a first portion of the environment from a video camera on a sharing device, (c) determining a registration data stream, where the registration data stream indicates a location and an orientation of the video stream within the still photo panorama of the environment; and (d) transmitting the video stream and the registration data stream to one or more viewing devices.

Abstract: Video data with high frame rates may be displayed on devices with limited resources (e.g., decoder and/or display resources). These devices may have their resources devoted to other tasks or may not be capable to display the video data at the high frame rates. The coding method may include coding the frames such that additional droppable frames are included in the encoded video data. The decoding method may include dropping droppable frames before the encoded video data is decoded to reduce the number of frames that will be decoded and displayed. These methods may be applied to video data that has a variable frame rate and may be combined with processing the image sequence for slow motion playback.

Abstract: A device configured to code video data includes: a memory configured to store video data, and at least one processor. The at least one processor is configured to: code information indicating whether a block from a current picture will flicker. A determination of whether the block from the current picture will flicker is based on the block in the current picture in a display order and a collocated block from a next picture in the display order.

Abstract: The quantization unit calculates the number of bits of location information determined based on the location of a coefficient level to be significant first in order of transmission and included in the image block, calculates the number of value information bits of a coefficient level to be significant, and sets a coefficient level to be significant in which the number of location information bits and the number of value information bits meet a predetermined condition to 0.