{"id":13481997,"url":"https://github.com/facebookresearch/fair_self_supervision_benchmark","last_synced_at":"2025-03-27T12:32:15.230Z","repository":{"id":49311944,"uuid":"186499279","full_name":"facebookresearch/fair_self_supervision_benchmark","owner":"facebookresearch","description":"Scaling and Benchmarking Self-Supervised Visual Representation Learning","archived":true,"fork":false,"pushed_at":"2021-10-12T21:34:03.000Z","size":52255,"stargazers_count":587,"open_issues_count":3,"forks_count":63,"subscribers_count":23,"default_branch":"main","last_synced_at":"2024-08-01T17:31:19.738Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"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/facebookresearch.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null}},"created_at":"2019-05-13T21:38:48.000Z","updated_at":"2024-06-26T02:16:00.000Z","dependencies_parsed_at":"2022-08-27T18:20:42.935Z","dependency_job_id":null,"html_url":"https://github.com/facebookresearch/fair_self_supervision_benchmark","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/facebookresearch%2Ffair_self_supervision_benchmark","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/facebookresearch%2Ffair_self_supervision_benchmark/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/facebookresearch%2Ffair_self_supervision_benchmark/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/facebookresearch%2Ffair_self_supervision_benchmark/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/facebookresearch","download_url":"https://codeload.github.com/facebookresearch/fair_self_supervision_benchmark/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":222252054,"owners_count":16955956,"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":[],"created_at":"2024-07-31T17:00:58.114Z","updated_at":"2024-10-30T15:31:40.159Z","avatar_url":"https://github.com/facebookresearch.png","language":"Python","funding_links":[],"categories":["Computer Vision","Python"],"sub_categories":["Image Representation Learning"],"readme":"## FAIR Self-Supervision Benchmark is deprecated. Please see [VISSL](https://vissl.ai/), a ground-up rewrite of benchmark in [PyTorch](https://pytorch.org/).\n\n# FAIR Self-Supervision Benchmark\n\nThis code provides various benchmark (and legacy) tasks for evaluating quality\nof visual representations learned by various self-supervision approaches. This code corresponds to our work on [Scaling and Benchmarking Self-Supervised Visual Representation Learning](https://arxiv.org/abs/1905.01235). The code is written in Python and can be used to **evaluate both PyTorch and Caffe2 models** (see [this](https://github.com/facebookresearch/fair_self_supervision_benchmark/tree/master/extra_scripts#using-pytorch-models)). We hope that this\nbenchmark release will provided a consistent evaluation strategy that will allow\nmeasuring the progress in self-supervision easily.\n\n## Introduction\nThe goal of fair_self_supervision_benchmark is to standardize the methodology for evaluating quality of visual representations learned by various self-supervision approaches. Further, it provides evaluation on a variety of tasks as follows:\n\n**Benchmark tasks**: The benchmark tasks are based on principle: a good representation (1) transfers to *many* different tasks, and, (2) transfers with *limited* supervision and *limited* fine-tuning. The tasks are as follows.\n- Image Classification\n  - [VOC07](http://host.robots.ox.ac.uk/pascal/VOC/pubs/everingham10.pdf)\n  - [COCO2014](https://arxiv.org/abs/1405.0312)\n  - [Places205](http://places.csail.mit.edu/places_NIPS14.pdf)\n- Low-Shot Image Classification\n  - [VOC07](http://host.robots.ox.ac.uk/pascal/VOC/pubs/everingham10.pdf)\n  - [Places205](http://places.csail.mit.edu/places_NIPS14.pdf)\n- Object Detection on [VOC07](http://host.robots.ox.ac.uk/pascal/VOC/pubs/everingham10.pdf) and [VOC07+12](http://host.robots.ox.ac.uk/pascal/VOC/pubs/everingham10.pdf) with frozen backbone for detectors:\n  - [Fast R-CNN](https://arxiv.org/abs/1504.08083)\n  - [Faster R-CNN](https://arxiv.org/abs/1506.01497)\n- [Surface Normal Estimation](https://web.eecs.umich.edu/~fouhey/2013/3dp/index.html)\n- Visual Navigation in [Gibson](https://arxiv.org/abs/1808.10654) Environment\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"demo/img-cls.png\" alt=\"Image Classification\" title=\"Image Classification\" width=\"200\" /\u003e\n  \u003cimg src=\"demo/obj-detection.png\" alt=\"Object Detection\" title=\"Object Detection\" width=\"200\" /\u003e\n  \u003cimg src=\"demo/surface-normal.png\" alt=\"Surface Normal Estimation\" title=\"Surface Normal Estimation\" width=\"200\" /\u003e\n  \u003cimg src=\"demo/visual-navigation.png\" alt=\"Visual Navigation\" title=\"Visual Navigation\" width=\"200\" /\u003e\n\u003c/p\u003e\n\nThese Benchmark tasks use the network architectures:\n\n- [AlexNet](https://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf)\n- [ResNet50](https://arxiv.org/abs/1512.03385)\n\n**Legacy tasks:** We also classify some commonly used evaluation tasks as legacy tasks for reasons mentioned in Section 7 of [paper](https://arxiv.org/abs/1905.01235). The tasks are as follows:\n\n- [ImageNet-1K](http://www.image-net.org/papers/imagenet_cvpr09.pdf) classification task\n- [VOC07](http://host.robots.ox.ac.uk/pascal/VOC/pubs/everingham10.pdf) full finetuning\n- Object Detection on [VOC07](http://host.robots.ox.ac.uk/pascal/VOC/pubs/everingham10.pdf) and [VOC07+12](http://host.robots.ox.ac.uk/pascal/VOC/pubs/everingham10.pdf) with full tuning for detectors:\n  - [Fast R-CNN](https://arxiv.org/abs/1504.08083)\n  - [Faster R-CNN](https://arxiv.org/abs/1506.01497)\n\n\n## License\n\nfair_self_supervision_benchmark is CC-NC 4.0 International licensed, as found in the LICENSE file.\n\n## Citation\n\nIf you use fair_self_supervision_benchmark in your research or wish to refer to the baseline results published in the [paper](https://arxiv.org/abs/1905.01235), please use the following BibTeX entry.\n\n```\n@article{goyal2019scaling,\n  title={Scaling and Benchmarking Self-Supervised Visual Representation Learning},\n  author={Goyal, Priya and Mahajan, Dhruv and Gupta, Abhinav and Misra, Ishan},\n  journal={arXiv preprint arXiv:1905.01235},\n  year={2019}\n}\n```\n\n## Installation\n\nPlease find installation instructions in [`INSTALL.md`](INSTALL.md).\n\n## Getting Started\n\nAfter installation, please see [`GETTING_STARTED.md`](GETTING_STARTED.md) for how to run various benchmark tasks.\n\n## Model Zoo\n\nWe provide models used in our [paper](https://arxiv.org/abs/1905.01235) in the [`MODEL_ZOO`](MODEL_ZOO.md).\n\n## References\n\n- [Scaling and Benchmarking Self-Supervised Visual Representation Learning](https://arxiv.org/abs/1905.01235). Priya Goyal, Dhruv Mahajan, Abhinav Gupta*, Ishan Misra*. Tech report, arXiv, May 2019.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ffacebookresearch%2Ffair_self_supervision_benchmark","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ffacebookresearch%2Ffair_self_supervision_benchmark","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ffacebookresearch%2Ffair_self_supervision_benchmark/lists"}