This work proposes a feature refined end-to-end tracking framework with a balanced performance using a high-level feature refine tracking framework. The feature refine module enhances the target feature representation power that allows the network to capture salient information to locate the target. The attention module is employed inside the feature refine mechanism to improve network discrimination power that augments the network ability to track the target in challenging scenarios.
This is an automatic object tracker project using image processing. This is the project done as a part of Instrumentation II course in fifth semester of Electronics and communication engineering, IOE Pulchowk Campus. In this project, a green ball is detected and followed by a vehicle. Members of this project are: Milan Maharjan (072bex446) Srijana Bhusal (072BEX440) Sunidhi Amatya (072bex443) Madhusudan Mainali (072BEX449)