A Framework for Evaluating Video Object Segmentation Algorithms

Abstract

Segmentation of moving objects in image sequences plays an important role in video processing and analysis. Evaluating the quality of segmentation results is necessary to allow the appropriate selection of segmentation algorithms and to tune their parameters for optimal performance. Many segmentation algorithms have been proposed along with a number of evaluation criteria. Nevertheless, no psychophysical experiments evaluating the quality of different video object segmentation results have been conducted. In this paper, a generic framework for segmentation quality evaluation is presented. A perceptually driven automatic method for segmentation evaluation is proposed and compared against an existing approach. Moreover, on the basis of subjective results, perceptual factors are introduced into the novel objective metric to meet the specificity of different segmentation applications such as video compression. Experimental results confirm the efficiency of the proposed evaluation criteria.