Video Bit Rate Adaptation via Sub-band Truncation and Motion Compensation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for adapting bit rates of Motion JPEG2000 compressed video streams to constrained transmission channels do not effectively minimize distortion, particularly in radio transmission, as they fail to account for error masking capabilities and temporal correlations between images.

Innovation Solution

A method that selects optimal truncation points and applies motion compensation to minimize distortion, using a key image to reconstruct images at higher resolutions while adhering to bit rate constraints, and optionally employs error-correcting codes to further reduce distortion from transmission errors.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If sub-bands are deleted to respect constrained bit rate, then bit rate constraint is satisfied, but image resolution and quality deteriorate

Engineering Contradiction:
Improvebit rateVSAvoidimage quality
Core Design Contradiction:
Quantity of substanceVSManufacturing precision

Solution Approach 1:

The compressed image is divided into multiple sub-bands representing different spatial resolutions. The method selectively transmits only the necessary sub-bands (e.g., low-frequency components) while omitting higher-frequency sub-bands, thereby segmenting the data to satisfy bit rate constraints while preserving essential image quality information.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different sub-bands are treated differently based on their importance. The method applies selective truncation where low-frequency sub-bands are retained for maintaining overall image structure, while high-frequency sub-bands are discarded. This local differentiation optimizes the balance between bit rate and image quality by prioritizing visually important components.

Inventive Principle:
Principle #3Local quality

2Reliability

If error-correcting codes are applied to increase total stream rate, then transmission reliability improves, but bit rate consumption increases

Engineering Contradiction:
Improvetransmission reliabilityVSAvoidbit rate
Core Design Contradiction:
ReliabilityVSQuantity of substance

Solution Approach 1:

Instead of applying error correction to the entire image stream, the method applies error-correcting codes selectively only to the most important sub-bands (e.g., low-frequency components) that carry essential image information. This partial application provides sufficient protection against transmission errors while minimizing the additional bit rate consumption.

Inventive Principle:
Principle #16Partial or excessive action

3Manufacturing precision

If temporal correlation is exploited for error masking, then image reconstruction quality improves, but complexity of processing increases

Engineering Contradiction:
Improveimage reconstruction qualityVSAvoidprocessing complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The method performs motion estimation and motion compensation in advance during the encoding stage, creating motion vectors and predicted blocks that can be used later for error masking. This preliminary preparation enables the decoder to reconstruct damaged blocks using temporal correlation without requiring complex real-time processing, thus improving reconstruction quality while controlling processing complexity.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2477402B1Method and device for adapting the bitrate of a compressed video stream transmitted in a constrained environment
Publication Date: 2014.09.24 THALES SA
  • EP2477402B1 patent drawingFigure 1
  • EP2477402B1 patent drawingFigure 2
  • EP2477402B1 patent drawingFigure 3

AI summary

Method and device for adapting the bit rate of a compressed video stream transmitted in a constrained environment. Method for adapting the bit rate of a video stream (300) comprising a plurality N of compressed images (Icn) at a plurality of spatial resolutions and/or a plurality of quality layers, said images being compressed separately from each other.