Video Encoding And Decoding With Filtered Virtual Reference Frames
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing video compression technologies face inefficiencies due to large data amounts in bitstreams and low compression rates, primarily because the similarity between reference frames and current images is low, leading to high residual values and reduced encoding efficiency.
Innovation Solution
The encoder and decoder sides determine a specific type of filter based on rendering information to improve the similarity between virtual reference frames and images, using mapping tables and interpolation to reduce residual values and data amounts in bitstreams.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If motion compensation and motion estimation are used to select reference frames, then the encoding process can be performed, but the similarity between reference frames and current image blocks is low, leading to large residual values and reduced encoding efficiency
Solution Approach 1:
The patent performs preliminary rendering to generate virtual reference frames before the encoding process. By pre-processing the reference frames through rendering operations (such as perspective correction, scaling, or other image processing), the system prepares higher-similarity reference images in advance, which are then used for motion compensation and prediction, ultimately reducing residual values and improving encoding efficiency
Solution Approach 2:
The patent introduces virtual reference frames as an intermediary between the original reference frames and the current image blocks. These virtual reference frames are generated through rendering operations and serve as a mediating layer that bridges the gap between existing reference frames and target image blocks, improving similarity and reducing prediction errors
2Reliability
If loop filtering is applied to generate reference frames, then image quality is improved, but the similarity between reference frames and current image blocks decreases, resulting in large differences and reduced compression performance
Solution Approach 1:
The patent applies loop filtering as a preliminary processing step to generate enhanced reference frames with improved image quality. By performing this filtering operation in advance during the rendering process, the system ensures that reference frames have better quality characteristics before being used for prediction, while the subsequent rendering operations maintain the similarity needed for good compression performance
Data Source
AI summary
A video encoding and decoding method and apparatus are disclosed, and relate to the computer field. An encoder side (a decoder side) selects a first type of filter from a plurality of specified types of filters based on a processing parameter indicated by rendering information, and obtains a virtual reference frame of to-be-processed data by using the first type of filter.