Multiple Object Tracking: A Review

Multiple Object Tracking (MOT) is an important computer vision task which has gained increasing attention due to its academic and commercial potential. Although different kinds of approaches have been proposed to tackle this problem, it still has many issues unsolved. For example, factors such as continuous appearance changes and severe occlusions result in difficulties for the task. In order to help the readers understand and learn this topic, we contribute a comprehensive and systematic review. We review the recent advances in various aspects about this topic and propose some interesting directions for future research. To our best knowledge, there has not been any review about this topic in the community. The main contribution of this review is threefold: 1) All key aspects in the multiple object tracking system, including what scenarios the researchers are working on, how their work can be categorized, what needs to be considered when developing a MOT system and how to evaluate a MOT system, are discussed in a clear structure. This review work could not only provide researchers, especially new comers to the topic of MOT, a general understanding of the state-of-the-arts, but also help them to comprehend the aspects of a MOT system and the inter-connected aspects. 2) Instead of listing and summarizing individual publications, we categorize the approaches in the key aspects involved in a MOT system. In each aspect, the methods are divided into different groups and each group is discussed in details for the principles, advances and drawbacks. 3) We provide some potential directions with insights for MOT, which are still open issues and need more research efforts. This would be helpful for researchers to identify further interesting problems.