Models & Raw results:
- Google Drive
- BaiduYun, code: qvfq
| Backbone | Pipeline | Dataset | A | R | EAO | FPS@GTX2080Ti | FPS@GTX1080Ti | Config. Filename | Model filename |
|---|---|---|---|---|---|---|---|---|---|
| AlexNet | Single template | VOT2018 | 0.588 | 0.243 | 0.373 | ~200 | ~185 | siamfcpp_alexnet.yaml | siamfcpp-alexnet-vot-md5_18fd31a2f94b0296c08fff9b0f9ad240.pkl |
| AlexNet | Simple multi-template strategy | VOT2018 | 0.597 | 0.215 | 0.370 | ~90 | ~75 | siamfcpp_alexnet-multi_temp.yaml | siamfcpp-alexnet-vot-md5_18fd31a2f94b0296c08fff9b0f9ad240.pkl |
| GoogLeNet | Single template | VOT2018 | 0.583 | 0.173 | 0.426 | ~80 | ~65 | siamfcpp_googlenet.yaml | siamfcpp-googlenet-vot-md5_f2680ba074213ee39d82fcb84533a1a6.pkl |
| GoogLeNet | Simple multi-template strategy | VOT2018 | 0.587 | 0.150 | 0.467 | ~50 | ~45 | siamfcpp_googlenet-multi_temp.yaml | siamfcpp-googlenet-vot-md5_f2680ba074213ee39d82fcb84533a1a6.pkl |
- The results reported in our paper were produced by the implement under the internal deep learning framework. Afterwards, we reimplement our tracking method under PyTorch and there could be some differences between the reported results (under internal framework) and the real results (under PyTorch).
- Differences in hardware configuration (e.g. CPU style / GPU style) may influence some indexes (e.g. FPS)
- Raw results here have been produced on a shared computing node equipped with Intel(R) Xeon(R) Gold 6130 CPU @ 2.10GHz and Nvidia GeForce RTX 2080Ti .
- "~" in the colomns for FPS denotes approximate values. FPS may vary due to factors other than code (e.g. hardware configuration / running status of machine).
- For VOT benchmark, models have been trained on ILSVRC-VID/DET, YoutubeBB, COCO, LaSOT, and GOT-10k (as described in our paper).