Psychovisually Optimized Dithering for Video Encoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for converting 10-bit pixel values to 8-bit values in image and video encoding often result in visible seams and distortion, particularly in areas with shallow gradients or subtle textures, due to inadequate dithering techniques that fail to preserve psychovisual characteristics.
Innovation Solution
Implementing a psychovisually optimized dithering mode that detects flat and non-flat regions in images or videos, dithering pixel values in non-flat regions while flattening those in flat regions, and biasing rounding towards full black or full white to maintain texture coherence and reduce bitrate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If standard dithering is applied during 10-bit to 8-bit conversion, then quantization errors are distributed, but visible seams and distortion appear in flat regions
Solution Approach 1:
The patent applies different dithering strategies to different image regions: flat regions use rounding to nearest integer to avoid noise and seams, while non-flat regions use dithering to preserve texture. This local differentiation resolves the contradiction by matching the processing method to the regional characteristics, eliminating visible artifacts while maintaining conversion accuracy.
Solution Approach 2:
The patent dynamically changes the dithering parameter (dither amount) based on the local image content. By detecting whether a region is flat or non-flat and adjusting the dithering intensity accordingly, the system achieves both accurate pixel conversion and avoidance of visible seams, resolving the technical contradiction.
2Reliability
If dithering is applied to preserve texture in non-flat regions, then psychovisual characteristics are maintained, but bitrate increases
Solution Approach 1:
The patent applies dithering only where necessary (non-flat regions with texture) and uses simple rounding in flat regions. This selective application maintains psychovisual characteristics in regions that need them while minimizing bitrate increase by avoiding unnecessary dithering in flat areas.
Solution Approach 2:
The patent applies partial dithering rather than full dithering across the entire image. By applying dithering only to the extent necessary for preserving texture in non-flat regions, the system maintains psychovisual fidelity while minimizing the bitrate penalty associated with increased complexity.
3Manufacturing precision
If rounding is biased towards full black or full white, then texture coherence is improved in low-luma areas, but encoding complexity increases
Solution Approach 1:
The patent changes the rounding parameter (bias towards full black or full white) based on the local luma level and texture characteristics. This adaptive parameter adjustment improves texture coherence in low-luma areas while managing encoding complexity through efficient detection and conditional application.
Data Source
AI summary
Techniques for psychovisually optimized dithering for image and video (e.g., frame) encoding are described. According to some embodiments, a computer-implemented method includes receiving a request to encode a video from a first depth of pixel values to a second, different depth of pixel values at a content delivery service, detecting a flat region without texturing and a non-flat region with texturing in a first frame of the video having the first depth of pixel values, dithering pixel values in the non-flat region and flattening pixel values in the flat region, encoding the dithered pixel values and the flattened pixel values of the frame into a first encoded frame having the second, different depth of pixel values by the content delivery service, and transmitting the first encoded frame from the content delivery service to a viewer device.