Temporal Block Merge Mode for Video Coding Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In video compression, existing technologies face challenges in efficiently representing object motion due to the need to describe arbitrary object shapes, which results in high bit usage and reduced coding efficiency, especially in scenarios with temporal redundancy.
Innovation Solution
The implementation of a method that determines and signals motion parameters for temporally-located blocks, allowing for the use of motion parameters from candidate blocks in a merge mode candidate list, including both spatially-located and temporally-located blocks, to reduce the need for redundant motion information transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If motion parameters are coded and transmitted for each individual block, then motion representation flexibility is improved, but bit usage increases and coding efficiency decreases
Solution Approach 1:
The patent merges the current block with a co-located block from a reference picture to form a combined region. Motion parameters are coded once for the combined region and applied to both blocks, eliminating redundant motion parameter transmission while maintaining motion representation flexibility through the merge mode indicator.
Solution Approach 2:
The patent creates a universal motion parameter set that serves multiple blocks simultaneously. The motion parameters coded for the combined region are universally applied to both the current block and the co-located block, reducing bit usage while maintaining adaptability through selective merge mode indication.
2Shape
If motion parameters are coded for each block in irregularly shaped objects, then object shape accuracy is improved, but coding efficiency decreases
Solution Approach 1:
The patent combines the current block with a co-located block from a reference picture into a single coding unit. This merging approach maintains accurate object shape representation by preserving block boundaries while improving coding efficiency by transmitting motion parameters once for the combined region rather than separately for each block.
3Quantity of substance
If spatial merge mode is used to merge current block with neighboring blocks, then bit usage is reduced, but temporal redundancy is not fully exploited
Solution Approach 1:
The patent extends the merge mode concept from spatial dimension to temporal dimension by introducing temporal merge mode. While spatial merge mode merges current block with neighboring blocks in the same picture, temporal merge mode merges current block with co-located block from reference picture, thereby exploiting temporal redundancy across different time points and further reducing bit usage.
Data Source
AI summary
In one embodiment, a method includes determining motion parameters for a temporally-located block to a current block of video content. The temporally-located block is located in a different picture from the current block. The temporally-located block is identified in a merge mode candidate list. The merge mode candidate list includes candidate blocks in which motion parameters are candidates to be used for the current block. The method then signals information to a decoder to indicate which motion parameters from a candidate block on the merge mode candidate list to use in decoding the current block. If a temporally-located block is identified, the method uses motion parameters for the temporally-located block in decoding the current block of video content.