- Google Drive
- Tencent Weiyun, code: wg47g7
If you want to run experiments on VOT, run the following script to compile region.
bash compile.sh
Set soft link to dataset directory (see config example)
ln -s path_to_datasets datasets
At path_to_datasets:
path_to_datasets
├── VOT # experiment configurations, in yaml format
│ ├── vot2018
│ │ ├── VOT2018
│ │ │ ├── ...
│ │ │ └── list.txt
│ │ └── VOT2018.json
│ └── vot2019
│ ├── VOT2019
│ │ ├── ...
│ │ └── list.txt
│ └── VOT2019.json
├── GOT-10k # same structure as full_data.zip
│ ├── train
│ ├── val
│ └── test
See DATA.md for a full description.
Auxilary files (list.txt / VOTXXXX.json) located at videoanalyst/evaluation/vot_benchmark/vot_list
Script for copying:
# Assuming that VOT benchmarks located under datasets/VOT and your
cd $(git rev-parse --show-toplevel) # change directory to the repo. root
cp videoanalyst/evaluation/vot_benchmark/vot_list/vot2018/VOT2018.json datasets/VOT/vot2018/
cp videoanalyst/evaluation/vot_benchmark/vot_list/vot2018/list.json datasets/VOT/vot2018/VOT2018/
We provide download links for VOT2018 / VOT2019:
- Google Drive
- Baidu Disk, code: xg4q
Acknowledgement:: Following datasets have been downloaded with TrackDat
- VOT2018
- VOT2019
As the official TrackingNet is hosted on Google Disk (with weekly download quota), we share a replica downloaded by us on BaiduDisk to facilitate others' life.
BaiduDisk, code: nkb9
Set soft link to model directory
ln -s path_to_models models
At path_to_models:
path_to_models
├── siamfcpp
│ ├── siamfcpp-alexnet-vot-md5_18fd31a2f94b0296c08fff9b0f9ad240.pkl
│ ├── siamfcpp-googlenet-vot-md5_f2680ba074213ee39d82fcb84533a1a6.pkl
│ ├── siamfcpp-tinyconv-vot-md5_cb9c2e8c7851ebf79677522269444cb2.pkl
│ ├── ...
│ ...
├── alexnet
│ └── alexnet-nopad-bn-md5_fa7cdefb48f41978cf35e8c4f1159cdc.pkl
├── googlenet
│ └── inception_v3_google-1a9a5a14-961cad7697695cca7d9ca4814b17a88d.pth
│
See SOT_TEST.md