Image Frame Segmentation for Independent Encoder Operation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In high-resolution image encoding, using two separate encoders to encode image frames divided into portions leads to challenges in motion vector search and synchronization, resulting in increased bit rate and reduced encoding quality due to the need for data copying and synchronization between encoders.
Innovation Solution
By estimating motion in the sequence of image frames, the image data of each frame is divided into two portions with a boundary positioned to minimize motion across it, allowing independent operation of the two encoders without data copying, thus reducing bit rate penalties and maintaining encoding quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If two separate encoders are used to encode high-resolution image frames, then encoding capacity and frame rate are improved, but synchronization complexity and data copying requirements increase
Solution Approach 1:
The image frame is divided into two separate portions (first portion and second portion), each encoded by a different encoder. This segmentation allows parallel encoding processing while maintaining independent encoder operation, reducing synchronization overhead compared to sharing a single encoder resource.
Solution Approach 2:
The reference buffer is extracted and duplicated for each encoder, with the first encoder having access to a first reference buffer and the second encoder having access to a second reference buffer. This eliminates the need for encoders to share or copy reference data, allowing independent operation without synchronization complexity.
2Measurement precision
If motion vector search is performed in the entire reference frame, then motion prediction accuracy is improved, but encoding time and frame rate decrease
Solution Approach 1:
The reference frame is segmented into multiple portions corresponding to the encoded portions, with each encoder searching only within its assigned portion. This reduces the search space from the entire reference frame to a smaller subset, maintaining sufficient motion prediction accuracy while enabling faster encoding to sustain high frame rates.
Solution Approach 2:
Each encoder is provided with a customized reference buffer containing only the reference data relevant to its encoded portion. This local optimization allows motion vector search to be confined to locally relevant areas, improving encoding speed while maintaining prediction accuracy for each specific region.
3Measurement precision
If reference buffer data is copied between encoders, then motion vector search accuracy is improved, but data transfer overhead and synchronization requirements increase
Solution Approach 1:
The reference buffer is extracted and instantiated separately for each encoder, eliminating the need for data copying between encoders. Each encoder has direct access to its own reference buffer, removing data transfer overhead and synchronization requirements while maintaining full access to necessary reference data for motion vector search.
Solution Approach 2:
Each encoder independently manages its own reference buffer, performing self-service without requiring data exchange with the other encoder. This autonomous operation eliminates inter-encoder data transfer overhead and synchronization complexity while maintaining encoding accuracy.
4Ease of operation
If search window is restricted to avoid reading from the other encoder's reference buffer, then encoder independence is improved, but motion prediction quality near boundaries deteriorates
Solution Approach 1:
The reference buffer is extracted and duplicated for each encoder, providing each with complete reference data independence. This allows encoders to operate fully independently without restricted search windows, as each has access to its own complete reference buffer, thereby maintaining encoding quality near boundaries while preserving encoder independence.
Solution Approach 2:
The solution moves from a single shared reference buffer dimension to multiple independent reference buffer dimensions (one per encoder). This dimensional expansion allows each encoder to access complete reference data without interference, eliminating boundary quality issues while maintaining independence.
Data Source
Figure 1~3
Figure 4~6
Figure 7~8
AI summary
The present invention relates to the field of image encoding. In particular, it relates to a method and device for encoding a plurality of image frames using two separate encoders, where each image frame is divided into two portions to each be encoded by one of the two encoders, where the image frame is divided to minimize motion across the boundary between the two portions, such that the two encoders may operate independently of each other without a substantial bit rate penalty or reduced encoding quality.