Video Coding Reference Sub-Region Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video coding standards require transmitting and processing entire reference frames for prediction, which is inefficient, especially when only a small portion of the static background has changed, leading to increased bandwidth usage and computational costs.
Innovation Solution
The method involves identifying and using a sub-region of the reference frame as a reference for prediction, which can be of a different size and location within the frame, allowing for efficient decoding and rescaling to match the current frame's size, thereby reducing the need for re-coding and re-transmission of the entire frame.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If entire reference frames are transmitted and processed for prediction, then prediction accuracy is maintained, but bandwidth usage and computational costs increase
Solution Approach 1:
The patent extracts only the necessary sub-region from the reference frame that contains relevant information for prediction, rather than transmitting and processing the entire frame. This extraction principle reduces the amount of data that needs to be transmitted and processed while maintaining prediction accuracy for the current frame's region of interest.
Solution Approach 2:
The reference frame is segmented into multiple regions, and only the relevant sub-region is selected and transmitted for prediction purposes. This segmentation allows the system to focus computational resources and bandwidth on the most important portions of the reference frame, improving efficiency without sacrificing prediction quality.
2Loss of energy
If sub-region of reference frame is used for prediction, then bandwidth and computational resources are reduced, but prediction flexibility must be maintained
Solution Approach 1:
The patent implements dynamic selection of sub-regions from reference frames, where the specific sub-region used for prediction can vary depending on the content and requirements of the current frame. This dynamic approach maintains prediction flexibility while consistently reducing bandwidth and computational resources compared to using entire frames.
Solution Approach 2:
The system changes parameters such as sub-region location, size, and selection criteria based on the specific prediction needs. By adjusting these parameters dynamically, the system maintains adaptability and versatility in prediction while operating with reduced data volumes, thereby conserving bandwidth and computational resources.
3Reliability
If reference frames are re-coded and re-transmitted, then prediction accuracy is ensured, but encoding and transmission time increase
Solution Approach 1:
The patent performs preliminary identification and extraction of relevant sub-regions from reference frames before the actual prediction process. By preparing and selecting the necessary sub-regions in advance, the system avoids the need for re-coding and re-transmission of entire frames, thereby reducing encoding and transmission time while maintaining prediction accuracy.
Solution Approach 2:
Instead of re-coding and re-transmitting entire reference frames, the system extracts and transmits only the necessary sub-regions. This extraction approach ensures that prediction accuracy is maintained for the relevant areas while significantly reducing the time required for encoding and transmission compared to processing complete frames.
Data Source
AI summary
A decoder includes circuitry configured to receive a bitstream, identify a first frame, locate a first independent reference region within the first frame, extract the first independent reference region from the first frame, decode a second frame using the first independent reference region as a reference for the second frame.