Object Detection Informed Video Coding for Bandwidth Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video coding systems face challenges in conserving bandwidth while maintaining image quality, especially in applications with limited and unpredictable communication bandwidth, such as smartphones and tablets, where existing codecs struggle to efficiently reduce bitrate without compromising image quality.
Innovation Solution
A video coder that detects objects within frames, adjusts coding parameters to provide high-quality coding for object regions and lower-quality coding for non-object regions, using block-based compression and pre-processing techniques like blurring filters to optimize compression efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If uniform high-quality coding is applied to the entire video frame, then image quality is preserved, but bandwidth consumption increases
Solution Approach 1:
The patent applies different coding qualities to different spatial regions within a video frame based on object detection results. High-quality coding is applied to regions containing detected objects while low-quality coding is applied to background regions, thereby preserving image quality where needed while reducing overall bandwidth consumption.
2Quantity of substance
If bitrate is reduced through compression, then bandwidth is conserved, but image quality deteriorates
Solution Approach 1:
The patent implements region-dependent compression where the degree of compression is adjusted locally based on the presence of detected objects. Regions containing objects undergo lighter compression to maintain quality, while background regions undergo heavier compression to achieve bitrate reduction, thus conserving bandwidth without uniformly sacrificing image quality.
3Productivity
If object-based selective coding is implemented, then coding efficiency improves, but system complexity increases
Solution Approach 1:
The patent segments the video frame into multiple regions based on object detection results, identifying foreground regions containing objects and background regions. This segmentation enables selective application of different coding parameters to different regions, improving coding efficiency while managing system complexity through structured region classification.
Data Source
AI summary
Embodiments of the present invention provide techniques for coding video data efficiently based on detection of objects within video sequences. A video coder may perform object detection on the frame and when an object is detected, develop statistics of an area of the frame in which the object is located. The video coder may compare pixels adjacent to the object location to the object's statistics and may define an object region to include pixel blocks corresponding to the object's location and pixel blocks corresponding to adjacent pixels having similar statistics as the detected object. The coder may code the video frame according to a block-based compression algorithm wherein pixel blocks of the object region are coded according to coding parameters generating relatively high quality coding and pixel blocks outside the object region are coded according to coding parameters generating relatively lower quality coding.