{"id":13737974,"url":"https://github.com/NVlabs/FreeSOLO","last_synced_at":"2025-05-08T15:32:08.563Z","repository":{"id":38884122,"uuid":"474775979","full_name":"NVlabs/FreeSOLO","owner":"NVlabs","description":"FreeSOLO for unsupervised instance segmentation, CVPR 2022","archived":false,"fork":false,"pushed_at":"2023-01-16T08:24:53.000Z","size":2308,"stargazers_count":314,"open_issues_count":14,"forks_count":32,"subscribers_count":4,"default_branch":"main","last_synced_at":"2024-11-15T06:32:57.336Z","etag":null,"topics":["cvpr2022","freesolo","instance-segmentation","object-detection","self-supervised-learning"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"other","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/NVlabs.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}},"created_at":"2022-03-27T22:33:03.000Z","updated_at":"2024-11-04T08:32:56.000Z","dependencies_parsed_at":"2023-02-10T02:01:43.384Z","dependency_job_id":null,"html_url":"https://github.com/NVlabs/FreeSOLO","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/NVlabs%2FFreeSOLO","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NVlabs%2FFreeSOLO/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NVlabs%2FFreeSOLO/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/NVlabs%2FFreeSOLO/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/NVlabs","download_url":"https://codeload.github.com/NVlabs/FreeSOLO/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":253096292,"owners_count":21853571,"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":["cvpr2022","freesolo","instance-segmentation","object-detection","self-supervised-learning"],"created_at":"2024-08-03T03:02:07.667Z","updated_at":"2025-05-08T15:32:07.469Z","avatar_url":"https://github.com/NVlabs.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"\n# FreeSOLO: Learning to Segment Objects without Annotations\n\nThis project hosts the code for implementing the FreeSOLO algorithm for unsupervised instance segmentation.\n\n\u003e [**FreeSOLO: Learning to Segment Objects without Annotations**](https://arxiv.org/abs/2202.12181),  \n\u003e Xinlong Wang, Zhiding Yu, Shalini De Mello, Jan Kautz, Anima Anandkumar, Chunhua Shen, Jose M. Alvarez  \n\u003e In: Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2022  \n\u003e *arXiv preprint ([arXiv 2202.12181](https://arxiv.org/abs/2202.12181))*  \n\n![pipeline](demo/pipeline.jpg)\n\n## Visual Results\n\n![vis](demo/vis.jpg)\n\n\n## Installation\n\n### Prerequisites\n\n- Linux or macOS with Python \u003e= 3.6\n- PyTorch \u003e= 1.5 and torchvision that matches the PyTorch installation.\n- scikit-image\n\n### Install PyTorch in Conda env\n\n```shell\n# create conda env\nconda create -n detectron2 python=3.6\n# activate the enviorment\nconda activate detectron2\n# install PyTorch \u003e=1.5 with GPU\nconda install pytorch torchvision -c pytorch\n```\n\n### Build Detectron2 from Source\n\nFollow the [INSTALL.md](https://github.com/facebookresearch/detectron2/blob/master/INSTALL.md) to install Detectron2 (commit id [11528ce](https://github.com/facebookresearch/detectron2/tree/11528ce083dc9ff83ee3a8f9086a1ef54d2a402f) has been tested).\n\n\n### Datasets\n\nFollow the [datasets/README.md](https://github.com/facebookresearch/detectron2/blob/master/datasets/README.md) to set up the MS COCO dataset.\n\n### Pre-trained model\n\nDownload the [DenseCL](https://github.com/WXinlong/DenseCL) pre-trained model from [here](https://cloudstor.aarnet.edu.au/plus/s/4sugyvuBOiMXXnC/download). Convert it to detectron2's format and put the converted model under \"training_dir/pre-trained/DenseCL\" directory.\n\n```shell\npython tools/convert-pretrain-to-detectron2.py {WEIGHT_FILE}.pth {WEIGHT_FILE}.pkl\n```\n\n\n\n## Usage\n\n### Free Mask\nDownload the prepared free masks in json format from [here](https://cloudstor.aarnet.edu.au/plus/s/AGcBm8cbqTWEKRS/download). Put it under \"datasets/coco/annotations\" directory. Or, generate it by yourself:\n\n```shell\nbash inference_freemask.sh\n```\n\n### Training\n\n```shell\n# train with free masks\nbash train.sh\n\n# generate pseudo labels\nbash gen_pseudo_labels.sh\n\n# self-train\nbash train_pl.sh\n```\n\n### Testing\nDownload the trained model from [here](https://cloudstor.aarnet.edu.au/plus/s/V8C0onE5H63x3RD/download).\n\n```shell\nbash test.sh {MODEL_PATH}\n```\nDisclaimer: This is a reimplementation and the numbers are slightly different from the original paper.\nThe above script will get instance segmentation results of 4.3 AP and 9.4 AP50. \n\n\n## Citations\nPlease consider citing our paper in your publications if the project helps your research. BibTeX reference is as follow.\n```BibTeX\n@article{wang2022freesolo,\n  title={{FreeSOLO}: Learning to Segment Objects without Annotations},\n  author={Wang, Xinlong and Yu, Zhiding and De Mello, Shalini and Kautz, Jan and Anandkumar, Anima and Shen, Chunhua and Alvarez, Jose M},\n  journal={arXiv preprint arXiv:2202.12181},\n  year={2022}\n}\n\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FNVlabs%2FFreeSOLO","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FNVlabs%2FFreeSOLO","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FNVlabs%2FFreeSOLO/lists"}