摘要:
Disclosed are various embodiments of three-dimensional (3D) video coding using scalable video coding (SVC) spatial scalability. In one embodiment, 3D video is encoded to generate a SVC base layer that includes a left first-resolution view and a right first-resolution view packed in a first frame. 3D video is encoded to generate a SVC enhancement layer that includes a left second-resolution view and a right second-resolution view packed in a second frame. The left second-resolution view and the right second-resolution view may have a higher spatial resolution than the left first-resolution view and the right first-resolution view.
摘要:
Methods and systems for hierarchically layered adaptive median motion vector smoothing are disclosed. Aspects of one method may include generating motion vectors (MVs) for video pictures at each level of a hierarchical motion estimation process, where each level may use different resolution video pictures. The MVs may be smooth filtered at each level to reduce or remove spurious MVs. The smooth filtering may comprise scalar median filtering and/or vector median filtering. The smooth filtering may receive as inputs a plurality of MVs, for example, the MV being filtered and eight MVs from the surrounding video blocks. The vector costs of the nine MVs may be compared to a threshold vector cost, and those MVs with costs above the threshold value may be discarded. The threshold value may be dynamically changed based on, for example, content of the video pictures and/or to adjust a window size of the smoothing filter.
摘要:
Aspects of a method and system for motion vector estimation using a pivotal pixel search are presented. Aspects of the system may include an image interpolation system that enables selection of an interpolated picture element neighborhood within an interpolated image frame. The image interpolation system may enable selection of one of a plurality of computed candidate motion vectors based on the location of the interpolated picture element neighborhood within the interpolated image frame. The image interpolation system may enable generation of picture element values within the selected interpolated picture element neighborhood based on at least the selected one of the plurality of computed candidate motion vectors.
摘要:
Aspects of a method and system for hierarchical motion estimation with multi-layer sub-pixel accuracy and motion vector smoothing are presented. Aspects of the system may include hierarchical motion vector computation that enables motion vectors to be computed at each level in the hierarchy based on a distinct pixel resolution level. A smoothing algorithm may be utilized to suppress spurious motion vector generation. The motion vectors computed at one level in the hierarchy may be utilized when computing motion vectors in a subsequent level. A bias value may be computed for each motion vector that provides an evaluation metric that may enable determination of whether the computed motion vector is to be utilized to enable generation of the interpolated image frame.
摘要:
A 3-dimensional (3D) video transmitter may be operable to encode a 3D video to generate a scalable video coding (SVC) base layer and a SVC enhancement layer. A first view such as a first high-resolution view and a second view such as a second high-resolution view of the 3D video in the SVC enhancement layer may be separate frames. A first half-resolution view and a second half-resolution view of the 3D video in the SVC base layer may be packed in a single frame. The first half-resolution view in the SVC base layer may be a base-layer reference for the first high-resolution view in the SVC enhancement layer for inter-layer prediction of spatial scalable coding. The first high-resolution view in the SVC enhancement layer may be an intra-layer reference for the second high-resolution view in the SVC enhancement layer for intra-layer prediction of temporal scalable coding.
摘要:
Certain aspects of a method and system for video compression with integrated picture rate up-conversion (PRUC) may include generating picture rate up-conversion (PRUC) data from received video data while the received video data is being encoded. The generated PRUC data may be encoded and communicated to a decoder in order to enable generation of a plurality of interpolated pictures. The generated encoded PRUC data may be communicated via a sideband to the decoder or by embedding the generated encoded PRUC data within the encoded received video data prior to the communication to the decoder.
摘要:
Aspects of a method and system for hierarchical motion estimation with multi-layer sub-pixel accuracy and motion vector smoothing are presented. Aspects of the system may include hierarchical motion vector computation that enables motion vectors to be computed at each level in the hierarchy based on a distinct pixel resolution level. A smoothing algorithm may be utilized to suppress spurious motion vector generation. The motion vectors computed at one level in the hierarchy may be utilized when computing motion vectors in a subsequent level. A bias value may be computed for each motion vector that provides an evaluation metric that may enable determination of whether the computed motion vector is to be utilized to enable generation of the interpolated image frame.
摘要:
Certain aspects of a method and system for motion-compensated picture rate up-conversion (PRUC) using information extracted from a compressed video stream may include extracting PRUC data from a compressed video data stream while the compressed video data stream is being decompressed by a video decompression engine. The PRUC data may comprise, for example, local block motion vectors, block coding modes, quantization levels, quantized residual data and decoded pictures. A plurality of interpolated pictures may be generated based on extracting the PRUC data.
摘要:
Certain aspects of a method and system for motion-compensated picture rate up-conversion (PRUC) of digital video using picture boundary processing may include generating one or more forward motion vectors and one or more backward motion vectors based on extracted picture rate up-conversion (PRUC) data. A cost of performing motion estimation of a particular block along the generated forward motion vectors and the generated backward motion vectors corresponding to the particular block may be calculated. The particular block may be a boundary block. A motion vector with the least cost may be selected and motion compensated to generate a plurality of interpolated pictures.
摘要:
A 3-dimensional (3D) video transmitter may be operable to encode a 3D video to generate a scalable video coding (SVC) base layer and a SVC enhancement layer. A first half-resolution view and a second half-resolution view of the 3D video in the SVC base layer may be packed in a first single frame. A first view such as a first high-resolution view and a second view such as a second high-resolution view of the 3D video in the SVC enhancement layer may be packed in a second single frame. The high-resolution may comprise a resolution that may be greater than half resolution. The first single frame in the SVC base layer may be used as a base-layer reference for the second single frame in the SVC enhancement layer for inter-layer prediction of spatial scalable coding.