Bit Rate Control for Video Coding via Selective CTU Processing

The video picture is divided into multiple coding tree units (CTUs), and each CTU is further divided into multiple blocks. The patent selectively processes only certain blocks within CTUs based on complexity metrics, rather than processing all smallest blocks uniformly. This segmentation approach maintains bit rate control precision for important regions while reducing overall computational complexity by skipping less important blocks.

The patent applies different processing strategies to different regions of the video based on their local characteristics. Blocks with higher complexity metrics undergo detailed coding parameter adjustment, while blocks with lower complexity metrics use simplified or skipped processing. This local quality approach ensures that computational resources are concentrated on regions that most impact bit rate control accuracy.

The patent segments the video processing into multiple passes or stages. In the first pass, complexity metrics are calculated for all blocks without full parameter adjustment. In subsequent passes, only selected blocks undergo complete coding parameter adjustment. This segmentation reduces the time-critical path while maintaining final bit rate control accuracy.

The patent performs preliminary calculations of complexity metrics and target bit allocations for all blocks before actual coding parameter adjustment. This preliminary action identifies which blocks require detailed processing and which can use simplified approaches, thereby reducing the total processing time while maintaining bit rate control accuracy for the most important blocks.

The patent divides the video coding process into multiple independent stages: complexity metric calculation, target bit allocation, and selective parameter adjustment. Each stage can be implemented as a separate hardware module, making the overall system more manageable and easier to implement in hardware compared to a monolithic approach that processes all blocks uniformly.

The patent implements partial action by selectively applying detailed coding parameter adjustment only to blocks that meet certain complexity criteria, while using simplified or default parameters for other blocks. This approach achieves acceptable video coding quality with significantly reduced hardware complexity compared to processing all blocks with full detail.

The patent segments the video picture into multiple CTUs and further into blocks, with each segment processed independently based on its complexity characteristics. This allows hardware resources to be dynamically allocated to different segments, reducing the peak resource requirements compared to processing all blocks simultaneously with full detail.

The patent applies different levels of processing quality to different spatial regions based on their local complexity metrics. High-complexity regions receive detailed parameter adjustment using more hardware resources, while low-complexity regions use simplified processing with fewer hardware resources. This local quality approach optimizes the overall hardware resource utilization while maintaining bit rate control precision where most needed.