Pytorch implementation of the Target-Aware Deep Tracking (TADT) method.
- Codes of the TADT tracker.
- Codes of visualization.
| tracker | OTB-50 | OTB2013 | OTB-100(OTB2015) |
|---|---|---|---|
| TADT-python | 0.615 | --- | 0.656 |
| TADT-official | --- | 0.680 | 0.660 |
rate: 77FPS (i7 8700k, RTX2080)
Note: We think that the tiny performance gap between TADT-python and TADT-official is caused by the difference between Matconvnet and pytorch
This code has been tested on Ubuntu 16.04, Python 3.7, Pytorch 1.1, CUDA 10, RTX 2080 GPU
numpy, cv2, matplotlib, scipy, yacs
- Clone the GIT repository:
$ git clone - Run the demo script to test the tracker:
python demo_tadt.py
Zikun Zhou Email: zikunzhou@163.com