Submit your paper : editorIJETjournal@gmail.com

Paper Title : Object Tracking Using Deep Reinforcement Learning

ISSN : 2395-1303
Year of Publication : 2022
10.5281/zenodo.7252092

Authors: Mr. Anil P Jawalkar, Sai Sangavi M, Sneha M, Meghana Rao M

         


Abstract
In this research, we offer an effective visual tracker that, through sequential actions honed using deep neural networks, directly captures a bounding box enclosing the target object in a video. Pre-trained using several training video sequences, the proposed deep neural network to govern tracking actions is then modified while actually tracking to allow for online response to a change in target and background. Deep reinforcement learning (RL) and supervised learning are both used for the pre-training. Even partially labelled data can be successfully used for semi-supervised learning when RL is applied. The suggested tracker is validated to attain a competitive performance at three times the speed of existing deep network-based trackers by the analysis of the object tracking benchmark data set.

Reference
[1] N.Wang and D -Y. Yeung, “learning a deep compact image representation for Visual tracking, “ in proc. Adv. Neural inf. Process. Syst., 2013, pp. 809-817. [2] J.F. Henriques, R. Caseiro, P. Martins, and J. Batista, “high-speed tracking with kernelized Correlation filters, “ IEEE trans. Pattern anal. Mach. Intell., Vol. 37, no. 3, pp. 583-596, mar.2015. [3] H. Grabner, C. Leistner, andH. Bischof, “Semi-supervised on-line boosting for robust Tracking, “ in Proc. Eur, Conf. Comput, Vis., 2008, pp, 234-247. [4] B. Babenko, M.H. Yang, and S. Belongie, “Robust object tracking with online multiple instance Learning. “ [5] M. Danellja, G. Hager, F. Khan, and M, Felsberg, “Accurate scale estimation for robust Visual tracking,” in Proc Brit, Mach. Vis. Conf., Nottingham, U.K., Sep. 2014. [6] ] J.F. Henriques, R. Caseiro, P. Martins, and J. Batista, “high-speed tracking with kernelized Correlation filters, “ IEEE trans. Pattern anal. Mach. Intell., Vol. 37, no. 3, pp. 583-596, mar.2015.
Keywords
— Object Tracking Using Deep Reinforcement Learning