Unified Scaling Differential Coding Bit Depth Extension
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing IBDI codecs face challenges in balancing coding gain from bit depth increase with adverse impacts on implementation complexity, particularly in efficiently compressing reference frames to minimize memory access and capacity issues.
Innovation Solution
The implementation of an improved IBDI codec that increases the bit depth of input pixels, uses adaptive offset calculations, and applies cyclic scanning patterns with delta coding to efficiently compress residual pixel values, optimizing bit usage and reducing encoding distortion through unified scaling and DPCM coding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the bit depth of input pixels is increased to improve video coding efficiency, then coding gain is achieved, but implementation complexity and memory requirements increase
Solution Approach 1:
The patent segments the bit depth increase into two distinct stages: an internal bit depth increase during encoding for improved prediction accuracy, and a separate bit depth adjustment stage during decoding to match the output requirements. This segmentation allows the system to benefit from higher precision computation while managing complexity through staged processing.
Solution Approach 2:
The patent applies preliminary bit depth scaling to reference frames before they are used in inter-frame prediction. By pre-scaling the reference frames to the target bit depth, the system improves prediction accuracy without requiring the entire decoding pipeline to operate at the higher bit depth, thus reducing overall implementation complexity.
2Quantity of substance
If reference frames are compressed to minimize memory access and capacity requirements, then memory efficiency is improved, but coding gain from bit depth increase is reduced
Solution Approach 1:
The patent applies different bit depth scaling strategies to different components of the video coding system. Reference frames undergo bit depth scaling to optimize their storage efficiency, while the prediction and transformation operations maintain higher precision through the internal bit depth increase. This local differentiation allows memory efficiency improvements without sacrificing coding gain in critical computation paths.
Solution Approach 2:
The patent performs bit depth scaling on reference frames in advance, before they are stored in memory buffers. This preliminary action ensures that reference frames are compressed to the appropriate bit depth for efficient storage, while the actual video encoding process benefits from the internal bit depth increase during prediction operations.
3Measurement precision
If internal bit depth increase is applied to improve inter-frame prediction, then prediction accuracy is enhanced, but memory access bandwidth requirements increase
Solution Approach 1:
The patent segments memory usage into two categories: reference frames stored at the target bit depth (lower precision) and prediction computations performed at increased internal bit depth (higher precision). This segmentation allows the system to maintain high prediction accuracy while reducing the memory bandwidth required for reference frame access, as the reference frames themselves occupy less memory space.
Solution Approach 2:
The patent applies bit depth scaling to reference frames before they are loaded into memory, reducing their memory footprint. This preliminary action decreases the amount of memory bandwidth required for subsequent access operations, while the internal bit depth increase during prediction ensures that prediction accuracy is maintained.
Data Source
AI summary
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.