Evaluation of Background Models with Synthetic and Real Data

As the other fields of computer vision [2, 19, 27], evaluating background subtraction (BS) algorithms is an important task to reveal the best approaches for a given application. Although the evaluation of BS techniques is an important issue, the impact of relevant papers that handle with both benchmarks and annotated dataset is limited [1, 6, 20]. Moreover, many authors that propose a novel approach compare their work with [25] or only a restricted part of the literature, but rarely with numerous recent related works.