Bi-directional Temporal Error Concealment for Video Frame Recovery

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing video decoding technologies struggle to effectively handle the loss of entire frames due to packet loss in video transmission, especially in low bit rate applications and high motion scenes, leading to severe distortions and error propagation.

Innovation Solution

The implementation of bi-directional temporal error concealment methods, which calculate bi-directional estimations for each pixel of a lost frame using both previous and next frame motion vectors, allowing for accurate reconstruction without increasing bit rate or delay.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If spatial-temporal predictive coding scheme is utilized to achieve high compression efficiency, then compression efficiency is improved, but error propagation to subsequent frames occurs

Engineering Contradiction:

Improvecompression efficiencyVSAvoiderror propagation

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent segments the error concealment process into independent pixel-level estimations rather than treating entire frames as units. By processing errors at the pixel level using bi-directional temporal estimation, the method prevents error propagation to subsequent frames while maintaining compression efficiency achieved through spatial-temporal predictive coding.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements feedback mechanisms by using both previous and next frame information to estimate lost pixels. The bi-directional temporal estimation uses feedback from both temporal directions (forward and backward) to reconstruct lost frames, thereby preventing error propagation while maintaining compression efficiency.

Inventive Principle:

Principle #23Feedback

2Loss of energy

If packet loss occurs in low bit rate applications, then transmission overhead is reduced, but entire frame loss occurs

Engineering Contradiction:

Improvetransmission overheadVSAvoidframe loss

Core Design Contradiction:

Loss of energyVSLoss of information

Solution Approach 1:

The patent applies partial action by performing error concealment at the pixel level rather than requiring complete frame recovery. By estimating each pixel independently using bi-directional temporal estimation, the system can recover partial information from lost frames without needing to transmit entire frames, thus reducing transmission overhead while minimizing information loss.

Inventive Principle:

Principle #16Partial or excessive action

3Manufacturing precision

If existing error concealment techniques are used to handle macroblock loss, then macroblock reconstruction is improved, but entire frame reconstruction fails

Engineering Contradiction:

Improvemacroblock reconstructionVSAvoidframe reconstruction

Core Design Contradiction:

Manufacturing precisionVSLoss of information

Solution Approach 1:

The patent transitions from block-based error concealment to pixel-based error concealment, adding a new dimension of precision. By operating at the pixel level rather than macroblock level, the system achieves both accurate macroblock reconstruction and complete frame reconstruction through bi-directional temporal estimation.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS7885339B2Bi-directional temporal error concealment

Publication Date: 2011.02.08 MICROSOFT TECHNOLOGY LICENSING LLC

AI summary

Systems and methods for bi-directional temporal error concealment are described. In one aspect, a lost frame is detected during encoded video decoding operations. Bi-directional estimations for each pixel of the lost frame are calculated to generate a current frame for bi-directional temporal error concealment of the lost frame.