Undergraduate Research Assistant since Sep. 2018
Table of Contents
Overview
Investigated correlation filter (CF)-based visual object tracking for unmanned aerial vehicles. By applying machine learning & deep learning techniques, we have improved the existing trackers on overall tracking performance in challenging scenarios with real-time operational capability.
Papers with code
Related work has been published in journals and conferences as follows:
Proposed a lightweight and generalizable triple attention strategy on CF-based framework by exploiting mutual independence of the appearance model and feature responses to implement real-time tracking for UAV.
🚩 Towards Robust Visual Tracking for Unmanned Aerial Vehicle with Tri-Attentional Correlation Filters in IROS 2020
Employed the adaptive GMSD-based context analysis and dynamic weighted filters for utilizing both contextual and historical information, and leveraged lightweight convolution features to efficiently raise the tracking robustness.
🚩 Robust Multi-Kernelized Correlators for UAV Tracking with Adaptive Context Analysis and Dynamic Weighted Filters in Neural Computing and Applications
Exploited the inter-frame information between prediction and backtracking phases for further incorporating the bidirectional incongruity error into the CF learning.
🚩 BiCF: Learning Bidirectional Incongruity-Aware Correlation Filter for Efficient UAV Object Tracking in ICRA 2020
