Abstract: A decoder includes circuitry configured to receive a bitstream, extract a header, determine, using the header, a global motion model, and decode a current block of a current frame using the global motion model. Related apparatus, systems, techniques and articles are also described.

Abstract: A method of identifying scleral reflections in an eye tracking system is disclosed. A glint is identified in an image from an image sensor, wherein the glint is a representation in the image of a reflection of light from a cornea of the eye of the user or from a sclera of the eye of the user. A first pixel intensity of the glint is determined, a second pixel intensity of neighbor pixels of the glint is determined, and an absolute value of the difference between the first pixel intensity of the glint and the second pixel intensity of the neighbor pixels of the glint is determined. The glint is identified as a representation of a reflection from the sclera of the eye of the user on condition that the determined absolute value of the difference is below a predetermine threshold value.

Abstract: A method includes obtaining respective filtered pixels for pixels of a reconstructed image; and obtaining an edge-preserved image using the respective filtered pixels. Obtaining the respective filtered pixels includes, for each pixel of the reconstructed image, obtaining a respective filtered pixel by selecting a pixel patch including the pixel and first neighboring pixels of the pixel that are at relative neighboring locations with respect to the pixel; calculating respective weights for the first neighboring pixels; and filtering the pixel using the respective weights of the first neighboring pixels and the neighboring pixels to obtain the respective filtered pixel.

Abstract: The present invention relates to a video encoding method, a video decoding method, and a device using the same, and the video encoding method according to the present invention comprises the steps of: specifying a tile and a slice by partitioning an inputted picture; performing encoding on the basis of the tile and the slice; and transmitting the encoded video information, wherein the picture is partitioned into one or more tiles and one or more slices, and the restrictions for parallel processing can be applied to the tiles and the slices.

Abstract: After initial intra prediction based on projection along an intra prediction direction, there may exist discontinuities between the predicted block and neighboring blocks. To smooth the block boundary, boundary filtering may be used with intra prediction. In particular, different methods may be used for boundary filtering. The particular filtering method to be used may be selected adaptively based on the intra prediction mode. When gradient-based filtering is selected, the gradient at a reference sample, that is at the same row or column as the current sample, in a reference layer is calculated as the difference between the reference sample and the predicted value for the reference sample using the intra prediction mode. Additional gradient at another reference sample from another reference layer may also be used. Then the initially predicted value for the sample may be adjusted based on the gradient(s) and a decay function.

Abstract: To enhance a mono-output-only controller such as a mobile OS to support selective mono/stereo/mixed output, a stereo controller is instantiated in communication with the mono controller. The stereo controller coordinates stereo output, but calls and adapts functions already present in the mono controller for creating surface and image buffers, rendering, compositing, and/or merging. For content designated for 2D display, left and right surfaces are rendered from a mono perspective; for content designated for 3D display, left and right surfaces are rendered from left and right stereo perspectives, respectively. Some, all, or none of available content may be delivered to a stereo display in 3D, with a remainder delivered in 2D, and with comparable content still delivered in 2D to the mono display. The stereo controller is an add-on; the mono controller need not be replaced, removed, deactivated, or modified, facilitating transparency and backward compatibility.

Abstract: Innovations in hash table construction and hash-based block matching for image encoding or video encoding are described. For example, an encoder determines hash values for base-size candidate blocks in a reference picture. The encoder stores, in a hash table, the hash values for the base-size candidate blocks. The encoder encodes a trial-size current block in a current picture. In some cases, the trial-size current block has a block size larger than the base block size. As part of the encoding, the encoder uses hash-based block matching, between base-size current blocks of the trial-size current block and the base-size candidate blocks, to identify a trial-size matching block, if any, in the reference picture. The encoder stores hash values only for the base-size candidate blocks. This can significantly reduce the computational cost and memory cost for hash table construction during encoding, without hurting compression efficiency or the overall speed of encoding.

Abstract: A camera calibration method includes obtaining a plurality of images of surroundings of a vehicle captured by a plurality of cameras, setting a region of interest (ROI) in each of the images, detecting one or more feature points of the set ROIs, matching a first feature point of a first ROI and a second feature point of a second ROI based on the detected feature points, calculating a first bird-view coordinate of the first feature point and a second bird-view coordinate of the second feature point, and calibrating the cameras by adjusting an extrinsic parameter of each of the cameras based on an error between the first bird-view coordinate and the second bird-view coordinate.

Abstract: Provided are embodiments including a computer-implemented method for performing recognition. The computer-implemented method includes receiving video data, and performing, at a pre-attention prediction module, a pre-attention prediction for the video data to generate first prediction priors. The computer-implemented method also includes receiving, at a dual attention module, data including the video data and data from the pre-attention prediction to generate attention maps, wherein the attention maps indicate a region of interest of a frame of the video data, wherein the dual attention module generates enhanced feature representations, and performing, at a post-attention prediction module, a post-attention prediction from data from the dual attention module based at least in part on the enhanced feature representation. Also provided are embodiments for a system and a computer program produce for performing recognition.

Abstract: Aspects of the subject disclosure may include, for example, a device that has a processing system including a processor; and a memory that stores executable instructions that, when executed by the processing system, facilitate performance of operations including selecting a downsampling technique for removing points from a point cloud in a volumetric video; selecting a data reduction percentage for a target bitrate for streaming the volumetric video; downsampling the point cloud in each frame of the volumetric video for the target bitrate using the downsampling technique and data reduction percentage selected, thereby creating a downsampled volumetric video; and compressing and encoding the downsampled volumetric video, thereby creating an encoded volumetric video. Other embodiments are disclosed.

Abstract: A video decoding method includes decoding a part of a bitstream to generate a decoded frame, and parsing at least one syntax element from the bitstream. The decoded frame is a projection-based frame that has projection faces packed in a cube-based projection layout. At least a portion of a 360-degree content of a sphere is mapped to the projection faces via cube-based projection. The at least one syntax element is indicative of packing of the projection faces in the cube-based projection layout.

Abstract: The present invention relates to a video encoding method, a video decoding method, and a device using the same, and the video encoding method according to the present invention comprises the steps of: specifying a tile and a slice by partitioning an inputted picture; performing encoding on the basis of the tile and the slice; and transmitting the encoded video information, wherein the picture is partitioned into one or more tiles and one or more slices, and the restrictions for parallel processing can be applied to the tiles and the slices.

Abstract: A server system obtains a first video sub-stream comprising a first plurality of images of a scene and obtains a second video sub-stream comprising a second plurality of images of at least a portion of the scene. Images of the second video sub-stream have a higher image resolution or are received at a higher frame rate than images of the first video sub-stream. The first video sub-stream is transmitted to a client device for display. A selection of a region of interest in the scene is received. The server systems locates image data of the selected region of interest from the second plurality of images, and provides the located image data of the selected region of interest to the client device for display.

Abstract: A method, a system and an archive server for determining an illumination setting for capturing an image of a component, belonging to a specific package type, in a pick-and-place machine are provided. The method comprises capturing a first image of a first component of the specific package type while the first component is illuminated by a first illumination component, and capturing a second image of the first component while the first component is illuminated by a second illumination component, the second illumination component being different from the first illumination component. An illumination setting may then be determined by creating a plurality of generated images based on the first and second image of the first component, and selecting the generated image that fulfils a predetermined quality measure.

Abstract: Systems, methods, and computing devices for capturing synthetic stereoscopic content are provided. An example computing device includes at least one processor and memory. The memory stores instructions that cause the computing device to receive a three-dimensional scene. The instructions may additionally cause the computing device to reposition vertices of the three-dimensional scene to compensate for variations in camera location in a directional stereoscopic projection and generate a stereoscopic image based on the repositioned vertices. An example method includes projecting a three-dimensional scene onto a left eye image cube and a right eye image cube and repositioning vertices of the three-dimensional scene to adjust for rendering from a single camera location. The method also includes mapping pixels of a stereoscopic image to points on the left eye image cube and the right eye image cube and generating the stereoscopic image using the values of the mapped pixels.

Abstract: Aspects of the disclosure provide a method and an apparatus for video coding. In some examples, an apparatus includes processing circuitry that divides a current block into a plurality of sub-blocks, derives a plurality of pieces of sub-block motion information for the plurality of sub-blocks, generates predicted sub-block samples for the sub-blocks according to the pieces of sub-block motion information, and generates a predicted block for the current block according to a combination of the predicted sub-block samples. The processing circuitry also selects a piece of the sub-block motion information from the pieces of sub-block motion information as representative motion information for the current block, and stores the representative motion information in a history-based motion vector prediction (HMVP) buffer.

Abstract: A portable photo studio system includes a portable photo studio and an electronic device such as a camera-enabled smartphone. The photo studio apparatus is configured to mount the electronic device for capturing images of a subject in the portable photo studio without a photographer. The electronic device not only captures images of the subject but also guides an image capture process by providing instructions to the subject for orienting their body. The electronic device is then used to transmit the images to an external computer system for processing.

Abstract: A variable bit rate generative compression method based on adversarial learning is provided. According to the method, a variance of a feature map of an encoding-decoding fill convolutional network is quantized to train a single generative model to perform variable bit rate compression. The method includes the following implementation steps of: constructing training and testing data sets through an image acquisition device; constructing a generative compression network based on an auto-encoder structure; according to a rate-distortion error calculation unit, alternately training a generative network; according to a target compression rate, calculating a mask threshold; based on a feature map channel redundancy index, calculating a mask; and performing lossless compression and decoding on the mask and the feature map. According to the invention, only a single model is trained, but compression results with different bit rates can be generated, and on a limit compression rate below 0.1 bpp.

Abstract: In one embodiment, an apparatus comprises a memory and a processor. The memory is to store sensor data captured by one or more sensors associated with a first device. Further, the processor comprises circuitry to: access the sensor data captured by the one or more sensors associated with the first device; determine that an incident occurred within a vicinity of the first device; identify a first collection of sensor data associated with the incident, wherein the first collection of sensor data is identified from the sensor data captured by the one or more sensors; preserve, on the memory, the first collection of sensor data associated with the incident; and notify one or more second devices of the incident, wherein the one or more second devices are located within the vicinity of the first device.

Abstract: A system and method for calibrating a forward-facing camera or sensor of a vehicle is provided. The system includes a calibration tool, the position of which can be adjusted on the vehicle to align a presentation screen of the calibration tool with the vehicle camera. A processor performs a method to determine the size and position of a calibration target to be displayed on the presentation screen, based on a position of the calibration tool relative to the vehicle camera and generates a calibration target image to be digitally transmitted to the presentation screen. The presentation screen displays the calibration target image at an appropriate height, angle, and distance from the camera or sensor being calibrated.