Abstract: Video processing methods and apparatuses for coding a current block comprise receiving input data of a current block, partitioning the current block into multiple sub-blocks, deriving sub-block MVs for the current block according to a sub-block motion compensation coding tool, constraining the sub-block MVs to form constrained sub-block MVs, and encoding or decoding the current block using the constrained sub-block MVs, and applying motion compensation to the current block using the constrained sub-block MVs to encode or decode the current block. The sub-block MVs may be constrained according to a size, width, or height of the current block or a sub-block, an inter prediction direction of one of control point MVs of the current block, the current block, or current sub-block, the control point MVs, or a combination of the above.

Abstract: Several techniques aimed at reducing computational complexity when encoding uses bi-predictively encoded frames (B-frames) are implemented in a video encoder. In an embodiment, B-frames are not used as reference frames for encoding P-frames and other B-frames. Non-use of B-frames allows a de-blocking filter used in the video encoder to be switched off when reconstructing encoded B-frames, and use of a lower complexity filter for fractional-resolution motion search for B-frames. In another embodiment, cost functions used in motion estimation for B-frames are simplified to reduce computational complexity. In one more embodiment, fractional pixel refinement in motion search for B-frames is simplified. In yet another embodiment, predictors used in motion estimation for a macro-block in a P-frame are selected from a B-frame that uses a same reference frame as the P-frame.

Abstract: Systems, methods, and apparatuses are described for processing video. Video content comprising a plurality of frames may be received. A viewing parameters associated with playback of the video content may be determined. One or more of luminance pixel data associated with a frame of the plurality of frames and chrominance pixel data associated with the frame is determined. A quantization matrix associated with the frame may be determined based on the viewing parameter and the one or more of luminance pixel data and chrominance pixel data.

Abstract: A method is provided that includes receiving pictures of a video sequence in a video encoder, and encoding the pictures to generate a compressed video bit stream that is transmitted to a video decoder in real-time, wherein encoding the pictures includes selecting a picture to be encoded as a delayed duplicate intra-predicted picture (DDI), wherein the picture would otherwise be encoded as an inter-predicted picture (P-picture), encoding the picture as an intra-predicted picture (I-picture) to generate the DDI, wherein the I-picture is reconstructed and stored for use as a reference picture for a decoder refresh picture, transmitting the DDI to the video decoder in non-real time, selecting a subsequent picture to be encoded as the decoder refresh picture, and encoding the subsequent picture in the compressed bit stream as the decoder refresh picture, wherein the subsequent P-picture is encoded as a P-picture predicted using the reference picture.

Abstract: The present disclosure relates to a method of motion segmentation (100) in a video stream. The method comprises the steps of: acquiring (101) a sequence of image frames; dividing (102) a first frame (401) into a plurality of image blocks (403); comparing (103) each image block (403) against a corresponding reference image block (404) and providing a measure of dissimilarity; for image blocks having a measure of dissimilarity less than a threshold: discarding (104a) the image blocks, and for image blocks having a measure of dissimilarity greater than the threshold: keeping (104b) the image blocks and further dividing the image blocks into a new plurality of image blocks (405); repeating the steps of dividing (102) and comparing (103) until a stop condition is met (105a); generating (106) a motion mask (407) indicating areas of movement (408).

Abstract: The present disclosure relates to methods and devices for image or frame processing including an apparatus, e.g., a GPU. In some aspects, the apparatus may identify a discontinuity between at least one first region of a first frame and at least one second region of a second frame, the at least one first region corresponding to the at least one second region. The apparatus may also determine whether the discontinuity between the at least one first region and the at least one second region is greater than or equal to a motion estimation threshold. The apparatus may also skip a motion estimation for the at least one first region and the at least one second region when the discontinuity between the at least one first region and the at least one second region is greater than or equal to the motion estimation threshold.

Abstract: The present invention provides a method, a device and a computer-readable medium for a pseudo seamless switch between different video sources and belongs to a field of the Internet technology. According to the method, a device and a computer-readable medium for a pseudo seamless switch between different video sources of the present invention, when a video source needs to be switched, the Web browser first establishes a target video source playback instance; determines a synchronization duration based on a playing state of an original video source and a network state; after the synchronization duration, the Web browser switches to the target video source playback instance to achieve seamless switching between various video sources, so as to effectively avoid blocking during playback and greatly improve user experience.