Identity assignment and retention needs multiple object detection and tracking. It plays a vital role in behavior analysis and gait recognition. The objective of Multiple Object Tracking (MOT) is to detect, track and retain identities from an image sequence. An occlusion is a major resistance in identity retention. It is a challenging task to handle occlusion while tracking varying number of person in the complex scene using a monocular camera. In MOT, occlusion remains a challenging task in real world applications. This paper uses Gaussian Mixture Model (GMM) and Hungarian Assignment (HA) for person detection and tracking. We propose an identity retention algorithm using Rotation Scale and Translation (RST) invariant feature descriptors. In addition, a segmentation based optimum demerge handling algorithm is proposed to retain proper identities under occlusion. The proposed approach is evaluated on a standard surveillance dataset sequences and it achieves 97 % object detection accuracy and 85% tracking accuracy for PETS-S2.L1 sequence and 69.7% accuracy as well as 72.3% precision for Town Centre Sequence.
Abstract:Tracking plays an important role to analyze the action of players, i.e movement of players and path corresponding to this movement. In recent years, the number of approaches to detecting players throughout the monocular image sequences has grown steadily. Severe occlusion makes this tracking challenging and standard algorithm fails to retain the identity of players. In soccer sequences, it is even more challenging due to erratic movement of players as well as players having clothes of the same color. Feature Descriptors do not require prior training to hold player's size, shape or color. Thus, this paper proposes feature descriptor based object tracking algorithm to tackle severe spatially extended and temporally short and long-term occlusions. The objective is to detect players and maintain their identities over the time even under the situation of intense occlusion. Gaussian Mixture Model (GMM) and Hungarian Assignment (HA) are used for player detection and association respectively. In the proposed approach, the feature descriptors are pre-calculated for the players having a reliable blob. In the case of demerge, these reliable feature descriptors are used to retain player identities. The proposed approach is evaluated on ISSIA Soccer dataset having six sequences having 25 frames per second (fps) for the image size of 1080 x 1920. The proposed method outperforms state of the art trackers by achieving 80% accuracy and 64% precision on ISSIA dataset.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.