Method and apparatus for coding of sample adaptive offset information
A method and apparatus for sharing context among different SAO syntax elements for a video coder are disclosed. Embodiments of the present invention apply CABAC coding to multiple SAO syntax elements according to a joint context model, wherein the multiple SAO syntax elements share the joint context. The multiple SAO syntax elements may correspond to SAO merge left flag and SAO merge up flag. The multiple SAO syntax elements may correspond to SAO merge left flags or merge up flags associated with different color components. The joint context model can be derived based on joint statistics of the multiple SAO syntax elements. Embodiments of the present invention code the SAO type index using truncated unary binarization, using CABAC with only one context, or using CABAC with context mode for the first bin associated with the SAO type index and with bypass mode for any remaining bin.
1. A method for Sample Adaptive Offset (SAO) processing of video data in a video decoder, the method comprising:
receiving a block of processed-reconstructed pixels associated with a picture from a media or a processor, wherein the block of processed-reconstructed pixels are decoded from a video bitstream;
determining a first SAO syntax element from the video bitstream, wherein the first SAO syntax element is decoded from the video bitstream using CABAC (context-based adaptive binary arithmetic coding) decoding,
wherein the first SAO syntax element corresponds to a SAO merge left flag and at least one second syntax element corresponds to a SAO merge up flag,
wherein the SAO merge left flag and the SAO merge up flag are associated with a first color component including one of a luma component and a chroma component, and
wherein the SAO merge left flag shares a joint context model with the SAO merge up flag; and
applying SAO processing to the block of processed-reconstructed pixels based on SAO information including the first SAO syntax element.
2. The method of claim 1 , wherein the joint context model is derived based on joint statistics of the first SAO syntax element and said at least one second syntax element.
3. The method of claim 1 , further comprising:
determining a SAO type index by parsing a coded SAO type index in the video bitstream, wherein the coded SAO type index is encoded according to truncated unary binarization.
4. A method for Sample Adaptive Offset (SAO) processing of video data in a video encoder, the method comprising:
receiving a block of processed-reconstructed pixels associated with a picture from a media or a processor;
determining a first SAO syntax element for the block of processed-reconstructed pixels;
applying CABAC (context-based adaptive binary arithmetic coding) encoding to the first SAO syntax element according to a joint context model,
wherein the first SAO syntax element corresponds to a SAO merge left flag and at least one second syntax element corresponds to a SAO merge up flag,
wherein the SAO merge left flag and the SAO merge up flag are associated with a first color component including one of a luma component and a chroma component, and
wherein the SAO merge left flag shares the joint context model with the SAO merge up flag; and
incorporating coded first SAO syntax element in a video bitstream associated with the video data.
5. The method of claim 4 , wherein the joint context model is derived based on joint statistics of the first SAO syntax element and said at least one second syntax element.
6. The method of claim 4 , further comprising:
determining a SAO type index for the block of processed-reconstructed pixels; and
incorporating a coded SAO type index in the video bitstream, wherein the coded SAO type index corresponds to the SAO type index being encoded according to truncated unary binarization.
7. An apparatus for Sample Adaptive Offset (SAO) processing of video data in a video decoder, the apparatus comprising:
at least one circuit configured for:
receiving a block of processed-reconstructed pixels associated with a picture from a media or a processor, wherein the block of processed-reconstructed pixels are decoded from a video bitstream;
determining a first SAO syntax element from the video bitstream, wherein the first SAO syntax element is decoded from the video bitstream using CABAC (context-based adaptive binary arithmetic coding) decoding,
wherein the first SAO syntax element corresponds to a SAO merge left flag and at least one second syntax element corresponds to a SAO merge up flag,
wherein the SAO merge left flag and the SAO merge up flag are associated with a first color component including one of a luma component and a chroma component, and
wherein the SAO merge left flag shares a joint context model with the SAO merge up flag; and
applying SAO processing to the block of processed-reconstructed pixels based on SAO information including the first SAO syntax element.
8. An apparatus for Sample Adaptive Offset (SAO) processing of video data in a video encoder, the apparatus comprising:
at least one circuit configured for:
receiving a block of processed-reconstructed pixels associated with a picture from a media or a processor;
determining a first SAO syntax element for the block of processed-reconstructed pixels;
applying CABAC (context-based adaptive binary arithmetic coding) encoding to the first SAO syntax element according to a joint context model,
wherein the first SAO syntax element corresponds to a SAO merge left flag and at least one second syntax element corresponds to a SAO merge up flag,
wherein the SAO merge left flag and the SAO merge up flag are associated with a first color component including one of a luma component and a chroma component, and
wherein the SAO merge left flag shares the joint context model with the SAO merge up flag; and
incorporating coded first SAO syntax element in a video bitstream associated with the video data.