Image Frame Encoding with Interest-Based Compression Limits
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for setting offset compression values in image frames result in artifacts or quality issues, especially in low contrast regions, and introduce high bitrate costs when reducing compression values uniformly across all regions.
Innovation Solution
A method and device that set offset compression values based on pixel block interest levels, selectively restricting these values to a predefined threshold in low contrast regions to avoid excessive reduction, ensuring optimal encoding without high bitrate costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If offset compression values are reduced by a fixed amount in all low contrast regions, then artifacts and quality issues are reduced, but bitrate cost increases significantly for pixel blocks with high interest levels
Solution Approach 1:
The patent applies local quality by differentiating the treatment of pixel blocks based on their interest levels. Pixel blocks with high interest levels (important regions) are excluded from the fixed amount reduction, while only pixel blocks with low interest levels undergo the reduction. This selective application ensures that video quality is improved in non-critical areas without sacrificing the quality of important regions, thereby avoiding excessive bitrate increase.
Solution Approach 2:
The patent changes the compression parameter (offset compression value) selectively based on the interest level of pixel blocks. Instead of applying a uniform parameter change across all low contrast regions, the system adjusts the compression value only for pixel blocks below a certain interest level threshold. This parameter differentiation resolves the contradiction by allowing quality improvement where acceptable and maintaining original compression settings where quality is critical.
2Manufacturing precision
If offset compression values are set low for all pixel blocks in low contrast regions, then quality issues are avoided, but encoding cost in terms of bitrate becomes very high
Solution Approach 1:
The patent implements local quality by setting low offset compression values only for pixel blocks with low interest levels within low contrast regions, while maintaining higher compression values for pixel blocks with high interest levels. This localized approach ensures that quality issues are avoided in non-critical areas without incurring excessive encoding costs for important regions, thereby improving encoding efficiency.
Solution Approach 2:
The system dynamically changes the compression parameter based on the interest level of each pixel block. By introducing an interest level threshold, the patent adjusts the offset compression value selectively: pixel blocks above the threshold retain higher compression values (lower encoding cost), while those below the threshold receive lower compression values (higher quality). This conditional parameter adjustment resolves the contradiction between quality and encoding efficiency.
Data Source
AI summary
An encoding device and a method for encoding an image frame comprising a plurality of pixel blocks are provided. A respective offset compression value is set for each of the plurality of pixel blocks based on an interest level associated with the pixel block, wherein each offset compression value defines an offset in relation to a reference compression value set for the image frame. In the image frame, one or more low contrast regions having a contrast which is below a predefined contrast threshold are identified. For pixel blocks within the one or more low contrast regions having set offset compression values based on the associated interest levels higher than a predefined offset compression threshold, the set offset compression values are selectively restricted to be at most equal to the predefined offset compression threshold. The image frame is then encoded using the set offset compression values.