Analytics-Modulated Video Compression for Surveillance Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing video compression techniques require custom encoders and decoders, making them costly and complex to deploy, and they do not efficiently allocate bits to regions of interest in surveillance video.
Innovation Solution
The method involves using analytics-modulated video compression, where encoding parameter values are adjusted based on the outputs of video analytics processes, allowing for dynamic adaptation of coding efficiency and prioritization of regions of interest.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If custom encoders and decoders are used to implement analytics-modulated video compression, then coding efficiency and region-of-interest fidelity are improved, but device complexity and deployment cost increase
Solution Approach 1:
The patent uses motion vectors and block data copied from reference frames to predict and reconstruct current frame regions. By copying motion information and applying it to generate predicted blocks, the system achieves efficient compression without requiring complex custom decoding hardware, as standard decoders can handle the copied data structures
Solution Approach 2:
The video frame is divided into multiple blocks that are processed independently with different encoding parameters. Regions of interest are segmented and allocated more bits while background regions use higher compression, improving overall coding efficiency without requiring a completely custom encoder architecture
2Device complexity
If uniform compression is applied to all video regions, then device complexity is reduced, but fidelity of regions of interest deteriorates
Solution Approach 1:
Different quantization parameters and compression settings are applied to different spatial regions within the same frame. Regions of interest use lower quantization (higher quality) while background regions use higher quantization (lower quality), achieving selective fidelity without requiring separate encoding pipelines for each region
Solution Approach 2:
The encoding parameters are dynamically adjusted based on detected motion and identified regions of interest. Motion vectors indicate which regions have changed, and encoding resources are dynamically reallocated to prioritize those regions, allowing the system to adapt compression quality to scene content rather than using static uniform compression
3Manufacturing precision
If high bit rate is used for all video frames, then video quality is maintained, but data transmission and storage requirements increase
Solution Approach 1:
The system changes encoding parameters (quantization parameters, bit allocation) based on scene analysis results. When motion is detected in certain regions, those regions receive higher bit allocation while other regions use aggressive compression, maintaining perceptual quality where needed while reducing overall data volume through parameter adaptation
Solution Approach 2:
Instead of applying high-quality encoding to the entire frame, the system applies high-quality encoding only to necessary regions (those with motion or identified as regions of interest). Background regions use lower quality encoding, achieving acceptable overall video quality with significantly reduced total bit rate by applying excessive quality only where necessary
Data Source
AI summary
A method and apparatus image analysis and motion detection using interframe coding, including, e.g., encoding surveillance video where one or more regions of interest are identified and the encoding parameter values associated with those regions are specified in accordance with intermediate outputs of a video analytics process. Such an analytics-modulated video compression approach allows the coding process to adapt dynamically based on the content of the surveillance images. In this manner, the fidelity of the region of interest is increased relative to that of a background region such that the coding efficiency is improved, including instances when no target objects appear in the scene. Better compression results can be achieved by assigning different coding priority levels to different types of detected objects.