We propose an unsupervised Dual-level Domain Adaptation TIR Tracking framework (DDAT), which can benefit from training on large-scale labeled RGB datasets and unlabeled TIR datasets. Specifically, to transfer the useful knowledge learned from the RGB dataset to TIR tracking, we first propose an adversarial-based adaptation module on both the semantic-level and the feature-level. While the semantic-level adaptation can reduce the semantic gap between the TIR and RGB tracking tasks, the feature-level adaptation can learn domain-invariant features for more robust tracking. Second, we propose a partial domain adaptation module to alleviate the negative transfer problem because the RGB and TIR tracking domains have non-identical class and feature spaces. Instead of aligning the entire feature space, this module adaptively selects partial similarity samples and features for alignment, thus obtaining more fine-grained aligned results. Third, we collect the currently largest-scale unlabeled TIR dataset to train the proposed framework.
-
You can download our trained models from Baidu Pan. Extraction Code: 1111
-
We provide a raw result of DDAT on the LSOTB-TIR100, LSOTB-TIR120, and PTB-TIR benchmarks in here. Extraction Code: 1111
- Clone the code and unzip it on your computer.
- Prerequisites: Ubuntu 22.04, Pytorch 2.2.2, GTX A100, CUDA 12.1.
- Download our trained models from here.
- Run
pysot_toolkit/test.pyto test a TIR sequence using the default model.
If you use the code or dataset, please consider citing our paper.
Feedback and comments are welcome!
Feel free to contact us via liuqiao.hit@gmail.com or liuqiao@stu.hit.edu.cn.