Awesome
DFAT
The official implementation of the Information Fusion paper "Exploring Fusion Strategies for Accurate RGBT Visual Object Tracking".
It is supported by the PRCI-Lab (https://github.com/PRCI-Lab)
Introduction
Based on the pysot, DFAT mainly focus on the bias caused by the different imaging mechanism.
<div align="center"> <img src="figs/bias.jpg" width="800px" /> <p>Illustration of databias</p> </div>Download models
The used model can be downloaded from baidu_disk. CODE:nfbe
Download testing datasets
Testing dataset like GTOT, RGBT210, RGBT234 can be downloaded from Chenglong Li's website
VOT-RGBT2019 and VOT-RGBT2020 datasets can be downloaded from baidu_disk. CODE:bcyo
Eval tracker
For VOT toolkit: python_siam_base.py
For testing: python test_gtot.py/test_vot.py
RFN
The RFN block is introduced from RFN-Nest