Unified scaling with differential coding for internal bit depth extension and reference frame compression
View Patent ↗A disclosed system, method and computer readable storage medium is provided to receive an input video comprising a plurality of video frames, each of which has a plurality of input pixels. Each input pixel is of n-bit depth (e.g., 8-bit depth). The method optionally increases the bitdepth of each input pixel by a predetermined factor (e.g., 4-bit factor). The method further compresses and reconstructs each pixels of the input video. The method further compresses and reconstructs each pixel of intermediate frame buffers stored in a decoder frame buffer. The method calculates an adaptive offset for each block (e.g., of size of 4×4) of reconstructed pixel values of the intermediate frame buffers. Furthermore, the method can compute two-sub-block offset values and uses them to compute an optimized adaptive offset to be applied to each pixel within the corresponding sub-block.
1. A method for improving coding efficiency of compression of an input video with internal bit depth increase (IBDI), the method comprising:
receiving the input video comprising a plurality of video frames, each video frame comprising a plurality of pixels, each pixel having a n-bit depth;
increasing the n-bit depth of each pixel by a predetermined factor;
compressing each pixel of the input video and reconstructing each compressed pixel of the input video;
calculating an adaptive offset for the value of each reconstructed pixel of the input video based on two sub-offset values, each of the sub-offset values determined using partial bits of total bits available for encoding the pixels; and
applying the adaptive offset to each reconstructed pixel values of the input video.
2. The method of claim 1 , wherein calculating the adaptive offset further comprises computing a fractional offset based on the fractional least significant bits of the pixel values, wherein the adaptive offset is set to the computed fractional offset.
3. The method of claim 1 , wherein calculating the adaptive offset further comprises constraining the adaptive offset value to a predetermined range of values.
4. The method of claim 1 , wherein each pixel of the input video is of 8-bit depth.
5. The method of claim 1 , wherein the predetermined factor is 4-bit depth.
6. The method of claim 1 , further comprising computing a residual pixel value for each pixel of the input video, wherein the residual pixel value is the difference between the original pixel value and a quantized minimum pixel value.
7. The method of claim 6 , further comprising scaling the residual pixel value based on a scaling factor.
8. The method claim 6 , further comprising computing a range of representable differences for the residual pixel values.
9. The method of claim 8 , wherein the range of representable differences is represented by an excess-N representation, wherein N is determined to maximize the range of representable differences.
10. A non-transitory computer-readable storage medium storing executable computer program instructions for improving coding efficiency of compression of an input video with internal bit depth increase, the computer program instructions comprising:
computer program code for receiving the input video comprising a plurality of video frames, each video frame comprising a plurality of pixels, each pixel having a n-bit depth;
computer program code for increasing the n-bit depth of each pixel by a predetermined factor;
computer program code for compressing each pixel of the input video and reconstructing each compressed pixel of the input video;
computer program code for calculating an adaptive offset for the value of each reconstructed pixel of the input video based on two sub-offset values, each of the sub-offset values determined using partial bits of total bits available for encoding the pixels; and
computer program code for applying the adaptive offset to each reconstructed pixel values of the input video.
11. The computer-readable storage medium of claim 10 , wherein the computer program code for calculating the adaptive offset further comprises computer program code for computing a fractional offset based on the fractional least significant bits of the pixel values, wherein the adaptive offset is set to the computed fractional offset.
12. The computer-readable storage medium of claim 10 , wherein the computer program code for calculating the adaptive offset further comprises computer program code for constraining the adaptive offset value to a predetermined range of values.
13. The computer-readable storage medium of claim 10 , further comprising computer program code for computing a residual pixel value for each pixel of the input video, wherein the residual pixel value is the difference between the original pixel value and a quantized minimum pixel value.
14. A computer system for improving coding efficiency of compression of a video with internal bit depth increase, the system comprising:
a non-transitory computer-readable storage medium storing executable computer program instructions for, the computer program instructions comprising:
computer program code for receiving the input video comprising a plurality of video frames, each video frame comprising a plurality of pixels, each pixel having a n-bit depth;
computer program code for increasing the n-bit depth of each pixel by a predetermined factor;
computer program code for compressing each pixel of the input video and reconstructing each compressed pixel of the input video;
computer program code for calculating an adaptive offset for the value of each reconstructed pixel of the input video based on two sub-offset values, each of the sub-offset values determined using partial bits of total bits available for encoding the pixels; and
computer program code for applying the adaptive offset to each reconstructed pixel values of the input video.
15. The computer system of claim 14 , wherein the computer program code for calculating the adaptive offset further comprises computer program code for computing a fractional offset based on the fractional least significant bits of the pixel values, wherein the adaptive offset is set to the computed fractional offset.
16. The computer system of claim 14 , wherein the computer program code for calculating the adaptive offset further comprises computer program code for constraining the adaptive offset value to a predetermined range of values.