AI Image Processing for High-Compression Video Quality Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image compression technologies, such as H.264/AVC and H.265/HEVC, struggle to maintain high-quality image transmission and storage with increasing data traffic demands, necessitating improved methods for high-definition image processing.
Innovation Solution
A method utilizing pre-processing techniques like frame skipping, down-scaling, and latent vector representation, combined with post-processing methods like frame interpolation and super-resolution, to enhance image quality while maintaining high compression efficiency, supported by artificial intelligence models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If standardized codec technologies (H.264/AVC, H.265/HEVC) are used for image compression, then compatibility and encoding efficiency are improved, but image quality degradation occurs and they become insufficient for high-definition streaming services
Solution Approach 1:
The patent applies pre-processing techniques (frame skipping, down-scaling) before encoding and post-processing techniques (frame interpolation, super-resolution) after decoding. This preliminary and subsequent action approach allows the system to work with lower-resolution or fewer frames during compression while restoring high quality afterward, thereby maintaining both compatibility with standard codecs and improving final image quality
Solution Approach 2:
The patent changes key parameters during processing: frame rate (skipping intermediate frames), resolution (down-scaling), and then reverses these changes through post-processing (frame interpolation to restore frame rate, super-resolution to restore detail). This parameter transformation approach enables high compression efficiency while maintaining the ability to restore original quality
2Productivity
If compression ratio is increased to reduce data traffic, then transmission efficiency is improved, but image quality deteriorates
Solution Approach 1:
The patent converts the harmful effect of compression artifacts and quality loss into a benefit by using post-processing AI techniques. The frame interpolation and super-resolution algorithms take the compressed, lower-quality intermediate frames and actively reconstruct enhanced quality output, turning the compression-induced degradation into an opportunity for quality enhancement through intelligent processing
3Productivity
If frame skipping is applied to reduce data amount, then compression efficiency is improved, but temporal resolution is reduced
Solution Approach 1:
The patent uses frame interpolation as an intermediary process. Instead of directly transmitting all original frames (high bandwidth) or using simple compression (low quality), the system skips frames during encoding but then uses AI-based interpolation to generate the missing intermediate frames during playback. This intermediary reconstruction process restores temporal resolution without requiring transmission of all original frames
4Quantity of substance
If down-scaling is used to reduce image size, then storage cost is reduced, but detail information is lost
Solution Approach 1:
The patent applies down-scaling as a preliminary compression step before encoding, reducing the amount of data that needs to be stored and transmitted. Then, as a compensatory post-processing step, super-resolution technology is applied to reconstruct the lost detail information. This two-stage approach (pre-compression down-scaling followed by post-enhancement super-resolution) allows efficient storage while recovering detail information
Data Source
AI summary
The present disclosure discloses an image processing method. The image processing method of the present disclosure may include obtaining image data including a plurality of image frames, performing preprocessing on the image data, encoding the preprocessed image data to generate encoded image data, and transmitting the encoded image data and information related to the preprocessing.