conference paper

Conference/ProceedingsCVPR 2006 Workshop (Perceptual Organization in Computer Vision POCV)
Start date17.06.2006
End date22.06.2006
AddressNew York
Author(s)Elisa Gelasca Drelie, Mustafa Karaman, Tourajd Ebrahimi, Thomas Sikora
TitleA Framework for Evaluating Video Object Segmentation Algorithms
AbstractSegmentation 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.
Key wordsvideo object, segmentation, perceptual metric, objective evaluation, subjective quality assessment,