[Paper Review] An Efficient Adaptive Boundary Matching Algorithm for Video Error Concealment

Seyed Mojtaba Marvasti-Zadeh, Hossein Ghanei-Yakhdan+1 more|arXiv (Cornell University)|Oct 24, 2016

Video Coding and Compression Technologies4 citations

TL;DR

This paper proposes an efficient adaptive boundary matching algorithm for video error concealment that improves motion vector recovery in damaged macroblocks by dynamically weighting boundary matching costs based on spatial and temporal context. By adaptively assigning higher weights to more reliable boundaries—especially when neighboring blocks are already concealed—it achieves significant PSNR gains (up to 5.88 dB) over existing methods with minimal computational overhead.

ABSTRACT

Sending compressed video data in error-prone environments (like the Internet and wireless networks) might cause data degradation. Error concealment techniques try to conceal the received data in the decoder side. In this paper, an adaptive boundary matching algorithm is presented for recovering the damaged motion vectors (MVs). This algorithm uses an outer boundary matching or directional temporal boundary matching method to compare every boundary of candidate macroblocks (MBs), adaptively. It gives a specific weight according to the accuracy of each boundary of the damaged MB. Moreover, if each of the adjacent MBs is already concealed, different weights are given to the boundaries. Finally, the MV with minimum adaptive boundary distortion is selected as the MV of the damaged MB. Experimental results show that the proposed algorithm can improve both objective and subjective quality of reconstructed frames without any considerable computational complexity. The average PSNR in some frames of test sequences increases about 5.20, 5.78, 5.88, 4.37, 4.41, and 3.50 dB compared to average MV, classic boundary matching, directional boundary matching, directional temporal boundary matching, outer boundary matching, and dynamical temporal error concealment algorithm, respectively.

Motivation & Objective

• To address motion vector recovery in video frames affected by data loss in error-prone networks such as the Internet and wireless systems.
• To improve both objective (PSNR) and subjective video quality after error concealment without increasing computational complexity.
• To develop a boundary matching method that adaptively evaluates candidate macroblocks based on reliability of their boundaries.
• To incorporate temporal and spatial context, especially when adjacent macroblocks are already concealed, by assigning different weights to boundaries.
• To outperform existing boundary matching techniques such as classic, directional, and temporal methods in error concealment performance.

Proposed method

• The algorithm uses outer boundary matching and directional temporal boundary matching to evaluate candidate macroblocks for damaged ones.
• It assigns adaptive weights to each boundary of candidate macroblocks based on the accuracy of the boundary and the concealment status of adjacent blocks.
• If adjacent macroblocks are already concealed, their boundaries are given higher weights to improve matching reliability.
• The motion vector with the minimum adaptive boundary distortion is selected as the final estimate for the damaged macroblock.
• The method dynamically combines spatial and temporal information to enhance matching accuracy while maintaining low computational cost.
• The algorithm operates at the decoder side, reconstructing lost motion vectors using neighboring block information and boundary consistency.

Experimental results

Research questions

• RQ1 How can motion vector recovery be improved in video error concealment under high packet loss conditions?
• RQ2 What weighting strategy can enhance boundary matching accuracy by leveraging spatial and temporal context?
• RQ3 Can adaptive boundary matching reduce distortion in reconstructed frames without increasing computational complexity?
• RQ4 How does the performance of the proposed method compare to classic, directional, and temporal boundary matching techniques?
• RQ5 To what extent does the concealment status of neighboring macroblocks affect the reliability of boundary matching?

Key findings

• The proposed algorithm improves average PSNR by 5.20 dB compared to the average motion vector method.
• It achieves a 5.78 dB PSNR gain over the classic boundary matching algorithm.
• The method outperforms directional boundary matching with a 5.88 dB improvement in PSNR.
• It provides a 4.37 dB PSNR gain over directional temporal boundary matching.
• The algorithm improves reconstruction quality by 4.41 dB compared to outer boundary matching.
• It achieves a 3.50 dB PSNR gain over the dynamical temporal error concealment algorithm.