EarlySkip Macroblock Encoding via Temporal Coherence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The computational intensity of motion estimation (ME) in H.264 video encoding is excessively high due to the complexity of macroblock encoding processes, making real-time encoding impractical.
Innovation Solution
The EarlySkip method reduces ME computation by leveraging temporal coherence between neighboring macroblocks, using a threshold value to early qualify skip prediction mode and thereby bypassing redundant processes, while maintaining image quality and coding efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If motion estimation is performed with multiple motion search block sizes and reference pictures in H.264 encoding, then compression efficiency is improved, but computational complexity increases significantly
Solution Approach 1:
The patent applies preliminary action by computing the skip prediction mode cost value (COSTMB_SKIP) before performing full motion estimation. This preliminary computation allows the encoder to early-qualify whether skip mode is suitable, avoiding unnecessary complex ME computations when skip mode is optimal, thus reducing overall computational complexity while maintaining compression efficiency
Solution Approach 2:
The patent segments the macroblock encoding process into two distinct paths: skip prediction mode evaluation and full motion estimation. By separating these processes and evaluating skip mode first, the patent avoids performing computationally intensive ME operations on macroblocks where skip mode is sufficient, thereby reducing device complexity while preserving compression efficiency
2Measurement precision
If full motion estimation is performed for every macroblock, then prediction accuracy is improved, but encoding speed decreases
Solution Approach 1:
The patent applies partial action by performing skip prediction mode evaluation (a partial ME process) on all macroblocks, and only performing full motion estimation on macroblocks where skip mode is not suitable. This selective approach maintains prediction accuracy where needed while accelerating encoding speed for macroblocks where simpler prediction suffices
Solution Approach 2:
By computing COSTMB_SKIP as a preliminary step before full ME, the patent enables early determination of whether skip mode is appropriate. This preliminary action filters out macroblocks that don't require full ME, thereby improving encoding speed without significantly compromising prediction accuracy for the remaining macroblocks
3Speed
If skip prediction mode is used to reduce computation, then encoding speed is improved, but image quality may deteriorate
Solution Approach 1:
The patent applies partial action by using skip prediction mode only for macroblocks where it is appropriate (determined by comparing COSTMB_SKIP against threshold T), and performing full motion estimation for other macroblocks. This selective application maintains image quality for complex regions while achieving speed improvements for simpler regions
Solution Approach 2:
The patent uses parameter changes by introducing a threshold parameter T that controls the decision between skip mode and full ME. By adjusting this parameter, the encoder can balance between encoding speed and image quality based on specific application requirements, allowing flexible optimization without sacrificing overall performance
Data Source
AI summary
The latest video-coding standards achieve higher coding efficiency than the previous video standards, while increasing the complexity and the difficulty of encoding. In a skip macroblock prediction mode some coding parameters (such as motion vectors and residual) are not coded. Selecting skip macroblock prediction mode reduces the size of the encoded bitstream while possibly deteriorating image quality. Previously the selection of the skip prediction mode is performed after motion estimation process. This invention determines whether each macroblock should be encoded in skip macroblock prediction mode before motion estimation. This invention substantially reduces computational cost with a very small deterioration in coding efficiency.