Adaptive Multi-Pass Video Encoding for Artifact Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video encoding technologies face challenges in balancing video quality and compression efficiency, particularly in reducing visible compression artifacts like banding and contouring, especially when encoding higher definition videos for diverse display sizes and network conditions.
Innovation Solution
Adaptive multi-pass risk-based video encoding using a neural network to assess the risk of compression artifacts in video segments and adjust encoding parameters accordingly, recommending less aggressive encoding for segments prone to artifacts and more aggressive encoding for segments with lower risk, thereby optimizing storage, transmission, and playback quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If higher strength of compression is applied to video data during encoding, then storage and transmission efficiency is improved, but visible compression artifacts increase and video quality deteriorates
Solution Approach 1:
The video content is divided into multiple segments or regions, and each segment is encoded with a different compression strength based on its visual importance. Critical regions with less detail or lower perceptual importance use higher compression, while important regions maintain lower compression to preserve quality.
Solution Approach 2:
Different portions of the video frame are assigned different quality levels based on local visual characteristics. Regions with fine details, high contrast, or semantic importance are encoded with higher quality (lower compression), while uniform or less important regions use lower quality (higher compression) settings.
2Manufacturing precision
If lower strength of compression is applied to video data during encoding, then video quality is improved, but storage and transmission efficiency deteriorates
Solution Approach 1:
The video is segmented into regions of varying importance, allowing the system to apply aggressive compression only to non-critical segments while maintaining high quality in important segments, thus achieving overall efficiency improvement without sacrificing essential video quality.
Solution Approach 2:
The encoding parameters such as quantization step size, bitrate, and compression ratio are dynamically adjusted based on the local visual content characteristics. This allows the system to optimize the balance between quality and efficiency for each specific region rather than using a uniform compression level throughout.
3Manufacturing precision
If higher definition video is encoded, then video quality is improved, but compression artifacts become more visible especially on larger displays
Solution Approach 1:
The encoding process identifies and protects regions that are more susceptible to visible artifacts, such as high-contrast edges, fine textures, and semantically important areas. These regions receive higher quality encoding with artifact-reduction techniques, while other regions use standard compression.
Solution Approach 2:
The system performs preliminary analysis of the video content to identify regions that are prone to compression artifacts before encoding. Pre-distortion or pre-processing techniques are applied to these identified regions to prevent artifact formation during the compression process.
4Productivity
If more aggressive encoding scheme is used, then data transmission and storage is optimized, but video quality and artifact reduction deteriorates
Solution Approach 1:
The video stream is divided into segments that are prioritized based on visual importance and network conditions. Critical segments are transmitted with higher quality encoding while less critical segments use more aggressive compression, enabling efficient bandwidth utilization without overall quality degradation.
Solution Approach 2:
The encoding scheme is made dynamic and adaptive, adjusting compression parameters in real-time based on network bandwidth availability, buffer status, and content characteristics. This allows the system to switch between aggressive and conservative encoding strategies to optimize both transmission efficiency and video quality under varying conditions.
Data Source
AI summary
Devices and methods are provided for adaptive multi-pass risk-based video encoding. A device may receive a segment of video frames encoded using first encoding parameters. The device may determine a group of pixels in a first video frame of the video frames. The device may determine characteristics associated with the group of pixels and may determine, based on the characteristics and a number of pixels in the group of pixels, a score associated with the segment, wherein the score is indicative of a visibility of banding compression artifact. The device may determine, based on the score, second encoding parameters associated with encoding the segment.