{"id":13443958,"url":"https://github.com/varunagrawal/tiny-faces-pytorch","last_synced_at":"2025-04-30T15:23:49.001Z","repository":{"id":34615795,"uuid":"143350153","full_name":"varunagrawal/tiny-faces-pytorch","owner":"varunagrawal","description":"Finding Tiny Faces in PyTorch","archived":false,"fork":false,"pushed_at":"2024-05-27T03:49:07.000Z","size":186,"stargazers_count":165,"open_issues_count":5,"forks_count":44,"subscribers_count":9,"default_branch":"master","last_synced_at":"2025-04-30T15:23:45.073Z","etag":null,"topics":["convolutional-neural-networks","deep-learning","face-detection","face-detection-application"],"latest_commit_sha":null,"homepage":null,"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/varunagrawal.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":"2018-08-02T22:18:56.000Z","updated_at":"2025-03-03T10:42:54.000Z","dependencies_parsed_at":"2024-10-28T05:53:57.748Z","dependency_job_id":"7b3fc6dd-e890-4a37-8986-fa852ff50838","html_url":"https://github.com/varunagrawal/tiny-faces-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/varunagrawal%2Ftiny-faces-pytorch","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/varunagrawal%2Ftiny-faces-pytorch/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/varunagrawal%2Ftiny-faces-pytorch/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/varunagrawal%2Ftiny-faces-pytorch/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/varunagrawal","download_url":"https://codeload.github.com/varunagrawal/tiny-faces-pytorch/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":251729853,"owners_count":21634296,"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":["convolutional-neural-networks","deep-learning","face-detection","face-detection-application"],"created_at":"2024-07-31T03:02:14.991Z","updated_at":"2025-04-30T15:23:48.977Z","avatar_url":"https://github.com/varunagrawal.png","language":"Python","readme":"# tiny-faces-pytorch\n\nThis is a PyTorch implementation of Peiyun Hu's [awesome tiny face detector](https://github.com/peiyunh/tiny). \n\nWe use (and recommend) **Python 3.6+** for minimal pain when using this codebase (plus Python 3.6 has really cool features).\n\n**NOTE** Be sure to cite Peiyun's CVPR paper and this repo if you use this code!\n\nThis code gives the following mAP results on the WIDER Face dataset:\n\n| Setting | mAP   |\n|---------|-------|\n| easy    | 0.902 |\n| medium  | 0.892 |\n| hard    | 0.797 |\n\n## Getting Started\n\n- Clone this repository.\n- Download the WIDER Face dataset and annotations files to `data/WIDER`.\n- Install dependencies with `pip install -r requirements.txt`.\n\nYour data directory should look like this for WIDERFace\n\n```\n- data\n    - WIDER\n        - README.md\n        - wider_face_split\n        - WIDER_train\n        - WIDER_val\n        - WIDER_test\n```\n\n## Pretrained Weights\n\nYou can find the pretrained weights which get the above mAP results [here](https://www.dropbox.com/scl/fi/md0lxok2uh2achx8r58mk/checkpoint_50.pth?rlkey=9y1acwj1k6c57tqck14t6as18\u0026dl=0).\n\n## Training\n\nJust type `make` at the repo root and you should be good to go!\n\nIn case you wish to change some settings (such as data location), you can modify the `Makefile` which should be super easy to work with.\n\n## Evaluation\n\nTo run evaluation and generate the output files as per the WIDERFace specification, simply run `make evaluate`. The results will be stored in the `val_results` directory.\n\nYou can then use the dataset's `eval_tools` to generate the mAP numbers (this needs Matlab/Octave).\n\nSimilarly, to run the model on the test set, run `make test` to generate results in the `test_results` directory.\n\n## Deployment\n\nTo run the model on your own image, please use the `detect_image.py` script.\nYou may have to adjust the probability and NMS thresholds to get the best results.\n","funding_links":[],"categories":["Python"],"sub_categories":[],"project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvarunagrawal%2Ftiny-faces-pytorch","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fvarunagrawal%2Ftiny-faces-pytorch","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fvarunagrawal%2Ftiny-faces-pytorch/lists"}