{"id":19098639,"url":"https://github.com/foolwood/dcfnet_pytorch","last_synced_at":"2025-04-09T23:21:33.321Z","repository":{"id":119091385,"uuid":"130957478","full_name":"foolwood/DCFNet_pytorch","owner":"foolwood","description":"DCFNet: Discriminant Correlation Filters Network for Visual Tracking","archived":false,"fork":false,"pushed_at":"2024-01-09T14:16:56.000Z","size":2615,"stargazers_count":208,"open_issues_count":15,"forks_count":60,"subscribers_count":11,"default_branch":"master","last_synced_at":"2025-04-02T16:49:43.522Z","etag":null,"topics":["cf","end-to-end-learning","fft","pytorch","tracking"],"latest_commit_sha":null,"homepage":"https://arxiv.org/pdf/1704.04057.pdf","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/foolwood.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2018-04-25T05:56:51.000Z","updated_at":"2024-09-16T17:53:07.000Z","dependencies_parsed_at":null,"dependency_job_id":"ed062d83-c12b-4483-8899-2e7a56b187dc","html_url":"https://github.com/foolwood/DCFNet_pytorch","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/foolwood%2FDCFNet_pytorch","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/foolwood%2FDCFNet_pytorch/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/foolwood%2FDCFNet_pytorch/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/foolwood%2FDCFNet_pytorch/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/foolwood","download_url":"https://codeload.github.com/foolwood/DCFNet_pytorch/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":248126376,"owners_count":21051910,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["cf","end-to-end-learning","fft","pytorch","tracking"],"created_at":"2024-11-09T03:46:37.194Z","updated_at":"2025-04-09T23:21:33.299Z","avatar_url":"https://github.com/foolwood.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# DCFNet_pytorch\u003csub\u003e([JCST](https://jcst.ict.ac.cn/en/article/doi/10.1007/s11390-023-3788-3))\u003c/sub\u003e\n\n[️‍🔥News️‍🔥] DCFNet is accepted in JCST. If you find [**DCFNet**](https://arxiv.org/pdf/1704.04057.pdf) useful in your research, please consider citing:\n\n```\n@Article{JCST-2309-13788,\ntitle = {DCFNet: Discriminant Correlation Filters Network for Visual Tracking},\njournal = {Journal of Computer Science and Technology},\nyear = {2023},\nissn = {1000-9000(Print) /1860-4749(Online)},\ndoi = {10.1007/s11390-023-3788-3},\t\nauthor = {Wei-Ming Hu and Qiang Wang and Jin Gao and Bing Li and Stephen Maybank}\n}\n```\n\n\n\nThis repository contains a Python *reimplementation* of the [**DCFNet**](https://arxiv.org/pdf/1704.04057.pdf).\n\n### Why implementation in python (PyTorch)?\n\n- Magical **Autograd** mechanism via PyTorch. Do not need to know the complicated BP.\n- Fast Fourier Transforms (**FFT**) supported by PyTorch 0.4.0. \n- Engineering demand. \n- Fast test speed (**120 FPS** on GTX 1060) and **Multi-GPUs** training.\n\n### Contents\n1. [Requirements](#requirements)\n2. [Test](#test)\n3. [Train](#train)\n4. [Citing DCFNet](#citing-dcfnet)\n\n## Requirements\n\n```shell\ngit clone --depth=1 https://github.com/foolwood/DCFNet_pytorch\n```\n\nRequirements for **PyTorch 0.4.0** and opencv-python\n\n```shell\nconda install pytorch torchvision -c pytorch\nconda install -c menpo opencv\n```\n\nTraining data (VID) and Test dataset (OTB).\n\n## Test\n\n```shell\ncd DCFNet_pytorch/track \nln -s /path/to/your/OTB2015 ./dataset/OTB2015\nln -s ./dataset/OTB2015 ./dataset/OTB2013\ncd dataset \u0026 python gen_otb2013.py\npython DCFNet.py\n```\n\n## Train\n\n1. Download training data. ([**ILSVRC2015 VID**](http://bvisionweb1.cs.unc.edu/ilsvrc2015/download-videos-3j16.php#vid)) \n\n   ```\n   ./ILSVRC2015\n   ├── Annotations\n   │   └── VID├── a -\u003e ./ILSVRC2015_VID_train_0000\n   │          ├── b -\u003e ./ILSVRC2015_VID_train_0001\n   │          ├── c -\u003e ./ILSVRC2015_VID_train_0002\n   │          ├── d -\u003e ./ILSVRC2015_VID_train_0003\n   │          ├── e -\u003e ./val\n   │          ├── ILSVRC2015_VID_train_0000\n   │          ├── ILSVRC2015_VID_train_0001\n   │          ├── ILSVRC2015_VID_train_0002\n   │          ├── ILSVRC2015_VID_train_0003\n   │          └── val\n   ├── Data\n   │   └── VID...........same as Annotations\n   └── ImageSets\n       └── VID\n   ```\n\n2. Prepare training data for `dataloader`.\n\n   ```shell\n   cd DCFNet_pytorch/train/dataset\n   python parse_vid.py \u003cVID_path\u003e  # save all vid info in a single json\n   python gen_snippet.py  # generate snippets\n   python crop_image.py  # crop and generate a json for dataloader\n   ```\n\n3. Training. (on multiple ***GPUs*** :zap: :zap: :zap: :zap:)\n\n   ```\n   cd DCFNet_pytorch/train/\n   CUDA_VISIBLE_DEVICES=0,1,2,3 python train_DCFNet.py\n   ```\n\n\n## Fine-tune hyper-parameter\n\n1. After training, you can simple test the model with default parameter.\n\n   ```shell\n   cd DCFNet_pytorch/track/\n   python DCFNet --model ../train/work/crop_125_2.0/checkpoint.pth.tar\n   ```\n\n2. Search a better hyper-parameter.\n\n   ```shell\n   CUDA_VISIBLE_DEVICES=0 python tune_otb.py  # run on parallel to speed up searching\n   python eval_otb.py OTB2013 * 0 10000\n   ```\n\n## Citing DCFNet\n\nIf you find [**DCFNet**](https://arxiv.org/pdf/1704.04057.pdf) useful in your research, please consider citing:\n\n```\n@article{wang2017dcfnet,\n  title={DCFNet: Discriminant Correlation Filters Network for Visual Tracking},\n  author={Wang, Qiang and Gao, Jin and Xing, Junliang and Zhang, Mengdan and Hu, Weiming},\n  journal={arXiv preprint arXiv:1704.04057},\n  year={2017}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ffoolwood%2Fdcfnet_pytorch","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ffoolwood%2Fdcfnet_pytorch","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ffoolwood%2Fdcfnet_pytorch/lists"}