Pseudo Two-Pass Video Decoder Memory Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video decoders face challenges in reducing memory requirements while maintaining acceptable picture quality, as methods like decimation in the spatial domain cause image blurring and decimation in the frequency domain result in unpredictable artefacts, especially in fast-moving sequences, and precise bit rate control is needed to avoid memory buffer overflow or degradation in picture quality.
Innovation Solution
A method employing pseudo two-pass decoding and one-pass encoding with a two-tiered rate control scheme, where the first pass extracts useful picture statistics for efficient one-pass encoding, allowing for significant memory reduction with minimal picture quality trade-offs, using techniques like Discrete Cosine Transform, quantization, and Variable-Length Coding, and re-encoding anchor frames to achieve a 10:1 memory reduction ratio.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If decimation in the spatial domain is used to reduce memory requirements, then memory size is reduced, but image quality deteriorates due to blurring
Solution Approach 1:
The patent divides the decoding process into two separate passes: a first pass that extracts essential picture statistics with minimal memory usage, and a second pass that performs full decoding for reference frames. This segmentation allows the system to obtain necessary decoding information without storing complete frame buffers, thus reducing memory size while maintaining image quality.
Solution Approach 2:
The first pass performs preliminary extraction of picture statistics (such as motion vectors, macroblock types, and DCT coefficients) before the actual decoding and display process. This preliminary action captures the essential information needed for subsequent processing without requiring full frame buffer storage, enabling memory reduction while preserving image quality.
2Quantity of substance
If decimation in the frequency domain with fixed bit rate is used, then memory requirements are reduced, but unpredictable artefacts appear especially in fast moving sequences
Solution Approach 1:
The patent implements dynamic bit rate control that adapts to the actual picture content and motion characteristics. The system adjusts the bit rate allocation based on the complexity of each macroblock and the motion activity detected during the first pass, ensuring consistent picture quality across different scene types including fast-moving sequences, while still achieving memory reduction through selective encoding.
Solution Approach 2:
The system changes the quantization parameters and bit rate allocation dynamically based on the picture statistics extracted in the first pass. By adjusting these parameters according to actual picture content rather than using a fixed bit rate, the system maintains reliable picture quality while achieving the desired memory reduction through optimized encoding in the second pass.
3Quantity of substance
If virtual-buffer-fullness control is used for bit rate control, then memory utilization can be managed, but picture quality degrades due to inaccurate bit rate variation control
Solution Approach 1:
The patent implements a feedback mechanism where the first pass analyzes the actual bit rate requirements and picture characteristics, and this information feeds into the second pass encoding process. The system uses this feedback to optimize the quantization parameters and bit allocation, ensuring that the target bit rate is achieved without excessive variation, thereby maintaining picture quality while managing memory utilization effectively.
Solution Approach 2:
The patent replaces the traditional virtual-buffer-fullness control mechanism with a direct picture-statistics-based control approach. Instead of relying on buffer fullness measurements that require tight control and cause quality degradation, the system directly uses the extracted picture statistics to guide bit rate allocation and encoding decisions, achieving both memory efficiency and picture quality without the need for aggressive buffer management.
Data Source
AI summary
A method of processing video frame data includes the steps of: receiving a video frame; partially decoding the video frame; fully decoding the video frame to produce macroblocks; determining video data parameters from the partially decoded video frame or both the partially and fully decoded video frame; and encoding the macroblocks based on the determined video data parameters to provide a compressed video frame for subsequent display.