P-Frame Video Encoding with Alternating Network Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In deep learning-based video compression/decompression, the quality of reconstructed images degrades over time due to accumulative errors, leading to a chain reaction that affects the compression performance of subsequent frames.
Innovation Solution
Employ multiple preset network parameter sets for encoding and decoding frames, alternating between high and low compression performance to reduce accumulative errors and improve overall compression efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single network parameter set is used for encoding multiple adjacent frames, then the encoding process is simple, but accumulative errors occur and compression performance degrades over time
Solution Approach 1:
The patent applies periodic action by alternating between different network parameter sets in a periodic manner. Specifically, different network parameter sets are selected based on frame identifiers (e.g., odd/even frame numbers), creating a periodic pattern where the encoding parameters change at regular intervals. This periodic switching prevents accumulative errors from propagating through all frames while maintaining a relatively simple encoding process.
2Reliability
If different network parameter sets are used for adjacent frames, then accumulative errors are reduced and compression performance is improved, but the encoding process complexity increases
Solution Approach 1:
The patent implements parameter changes by switching between different network parameter sets for encoding adjacent frames. Each network parameter set contains different parameters (such as convolutional layer parameters, activation functions, or other neural network configuration parameters) that are optimized for different compression performance requirements. This parameter variation allows the system to reduce accumulative errors and improve compression performance while managing encoding complexity through systematic parameter selection.
Data Source
AI summary
An encoding method includes obtaining a to-be-encoded frame, where the to-be-encoded frame is a P-frame, determining, from M preset network parameter sets, a network parameter set corresponding to the to-be-encoded frame, where the M preset network parameter sets respectively correspond to different compression performance information, and M is an integer greater than one, and encoding, by an encoding network, and based on the network parameter set corresponding to the to-be-encoded frame, the to-be-encoded frame to obtain a bitstream representative of the to-be-encoded frame.