Object-Based Parametric Video Encoder for Perceptual Redundancy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video compression techniques do not efficiently address perceptual redundancy in background texture areas and regions of interest (ROI), leading to suboptimal compression efficiency.
Innovation Solution
An object-based parametric compression method that segments frames into object and background regions, using orthonormal basis projection coefficients for objects and auto-regressive modeling for backgrounds, allowing for separate and prioritized coding to achieve higher compression rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If standard video compression techniques (predictive coding, transform coding, vector quantization) are used, then inter-pixel redundancy is removed through motion compensation and transform-based approaches, but perceptual redundancy in background texture areas and regions of interest is not efficiently addressed
Solution Approach 1:
The video sequence is segmented into multiple regions of interest (ROIs) and background areas based on perceptual importance. The segmentation module divides each frame into foreground objects and background regions, allowing different compression strategies to be applied to different segments. This enables efficient exploitation of perceptual redundancy by treating significant regions differently from less important background areas.
Solution Approach 2:
Different compression qualities and techniques are applied to different regions of the video frame. Regions of interest receive higher quality encoding with more bits allocated, while background texture areas use more aggressive compression. This local quality differentiation maximizes perceptual efficiency by preserving important visual information while compressing less critical areas.
2Ease of manufacture
If uniform compression is applied to all regions, then implementation is simple, but different priorities cannot be assigned to different objects and background regions
Solution Approach 1:
The system dynamically adjusts compression parameters and bit-rate allocation based on the temporal and spatial significance of different regions. Motion analysis and change detection modules continuously assess which regions require higher quality encoding, allowing the system to adapt compression strategies frame-by-frame and region-by-region rather than using static uniform compression.
Solution Approach 2:
Compression parameters such as quantization step size, transform block size, and motion search range are changed dynamically based on region importance. Regions of interest use finer quantization and larger transform blocks for better quality, while background areas use coarser quantization. Bit-rate allocation parameters are adjusted to prioritize significant regions, enabling flexible prioritization without complex manual configuration.
Data Source
AI summary
A video compression framework based on parametric object and background compression is proposed. At the encoder, an object is detected and frames are segmented into regions corresponding to the foreground object and the background. The encoder generates object motion and appearance parameters. The motion or warping parameters may include at least two parameters for object translation; two parameters for object scaling in two primary axes and one object orientation parameter indicating a rotation of the object. Particle filtering may be employed to generate the object motion parameters. The proposed methodology is the formalization of the concept and usability for perceptual quality scalability layer for Region(s) of Interest. A coded video sequence format is proposed which aims at “network friendly” video representation supporting appearance and generalized motion of object(s).