{"id":13546264,"url":"https://github.com/wkentaro/fcn","last_synced_at":"2025-04-05T03:05:10.832Z","repository":{"id":51195981,"uuid":"57879235","full_name":"wkentaro/fcn","owner":"wkentaro","description":"Chainer Implementation of Fully Convolutional Networks. (Training code to reproduce the original result is available.)","archived":false,"fork":false,"pushed_at":"2021-10-06T19:49:23.000Z","size":20953,"stargazers_count":217,"open_issues_count":1,"forks_count":97,"subscribers_count":14,"default_branch":"main","last_synced_at":"2025-03-29T02:09:00.893Z","etag":null,"topics":["chainer","computer-vision","convolutional-networks","deep-learning","fcn","fcn8s","segmentation","semantic-segmentation"],"latest_commit_sha":null,"homepage":"","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/wkentaro.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":"2016-05-02T09:44:16.000Z","updated_at":"2025-01-26T07:53:19.000Z","dependencies_parsed_at":"2022-09-19T04:30:43.418Z","dependency_job_id":null,"html_url":"https://github.com/wkentaro/fcn","commit_stats":null,"previous_names":[],"tags_count":73,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/wkentaro%2Ffcn","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/wkentaro%2Ffcn/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/wkentaro%2Ffcn/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/wkentaro%2Ffcn/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/wkentaro","download_url":"https://codeload.github.com/wkentaro/fcn/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247280262,"owners_count":20912967,"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":["chainer","computer-vision","convolutional-networks","deep-learning","fcn","fcn8s","segmentation","semantic-segmentation"],"created_at":"2024-08-01T12:00:34.879Z","updated_at":"2025-04-05T03:05:10.817Z","avatar_url":"https://github.com/wkentaro.png","language":"Python","funding_links":[],"categories":["Preferred Networks Research"],"sub_categories":["Computer Vision"],"readme":"fcn - Fully Convolutional Networks\n==================================\n\n[![PyPI Version](https://img.shields.io/pypi/v/fcn.svg)](https://pypi.python.org/pypi/fcn)\n[![Python Versions](https://img.shields.io/pypi/pyversions/fcn.svg)](https://pypi.org/project/fcn)\n[![GitHub Actions](https://github.com/wkentaro/fcn/workflows/CI/badge.svg)](https://github.com/wkentaro/fcn/actions)\n\nChainer implementation of [Fully Convolutional Networks](https://github.com/shelhamer/fcn.berkeleyvision.org).\n\n\nInstallation\n------------\n\n```bash\npip install fcn\n```\n\n\nInference\n---------\n\nInference is done as below:\n\n```bash\n# forwaring of the networks\nimg_file=https://farm2.staticflickr.com/1522/26471792680_a485afb024_z_d.jpg\nfcn_infer.py --img-files $img_file --gpu -1 -o /tmp  # cpu mode\nfcn_infer.py --img-files $img_file --gpu 0 -o /tmp   # gpu mode\n```\n\n\u003cimg src=\".readme/fcn8s_26471792680.jpg\" width=\"80%\" \u003e\n\nOriginal Image: \u003chttps://www.flickr.com/photos/faceme/26471792680/\u003e\n\n\nTraining\n--------\n\n```bash\ncd examples/voc\n./download_datasets.py\n./download_models.py\n\n./train_fcn32s.py --gpu 0\n# ./train_fcn16s.py --gpu 0\n# ./train_fcn8s.py --gpu 0\n# ./train_fcn8s_atonce.py --gpu 0\n```\n\nThe accuracy of original implementation is computed with (`evaluate.py`) after converting the caffe model to chainer one\nusing `convert_caffe_to_chainermodel.py`.\\\nYou can download vgg16 model from here: [`vgg16_from_caffe.npz`](https://drive.google.com/uc?id=1lKjYpFOc9EAPIt7foOO2nbaOWIpIZRaw).\n\n**FCN32s**\n\n| Implementation | Accuracy | Accuracy Class | Mean IU | FWAVACC | Model File |\n|:--------------:|:--------:|:--------------:|:-------:|:-------:|:----------:|\n| [Original](https://github.com/shelhamer/fcn.berkeleyvision.org/tree/main/voc-fcn32s) | 90.4810 | 76.4824 | 63.6261 | 83.4580 | [`fcn32s_from_caffe.npz`](https://drive.google.com/uc?id=1l5Ubj7zflQAwaNI7G35DzNcMoTf8CDLW) |\n| Ours (using `vgg16_from_caffe.npz`) | **90.5668** | **76.8740** | **63.8180** | **83.5067** | - |\n\n**FCN16s**\n\n| Implementation | Accuracy | Accuracy Class | Mean IU | FWAVACC | Model File |\n|:--------------:|:--------:|:--------------:|:-------:|:-------:|:----------:|\n| [Original](https://github.com/shelhamer/fcn.berkeleyvision.org/tree/main/voc-fcn16s) | 90.9971 | **78.0710** | 65.0050 | 84.2614 | [`fcn16s_from_caffe.npz`](https://drive.google.com/uc?id=1kwLpLz1jzQqqF5QxQphsO4-aObQIQMWS) |\n| Ours (using `fcn32s_from_caffe.npz`) | 90.9671 | 78.0617 | 65.0911 | 84.2604 | - |\n| Ours (using `fcn32s_voc_iter00092000.npz`) | **91.1009** | 77.2522 | **65.3628** | **84.3675** | - |\n\n**FCN8s**\n\n| Implementation | Accuracy | Accuracy Class | Mean IU | FWAVACC | Model File |\n|:--------------:|:--------:|:--------------:|:-------:|:-------:|:----------:|\n| [Original](https://github.com/shelhamer/fcn.berkeleyvision.org/tree/main/voc-fcn8s) | 91.2212 | 77.6146 | 65.5126 | 84.5445 | [`fcn8s_from_caffe.npz`](https://drive.google.com/uc?id=1l_5RK28JRL19wpT22B-DY9We3TVXnnQQ) |\n| Ours (using `fcn16s_from_caffe.npz`) | 91.2513 | 77.1490 | 65.4789 | 84.5460 | - |\n| Ours (using `fcn16s_voc_iter00100000.npz`) | **91.2608** | **78.1484** | **65.8444** | **84.6447** | - |\n\n**FCN8sAtOnce**\n\n| Implementation | Accuracy | Accuracy Class | Mean IU | FWAVACC | Model File |\n|:--------------:|:--------:|:--------------:|:-------:|:-------:|:----------:|\n| [Original](https://github.com/shelhamer/fcn.berkeleyvision.org/tree/main/voc-fcn8s-atonce) | **91.1288** | **78.4979** | **65.3998** | **84.4326** | [`fcn8s-atonce_from_caffe.npz`](https://drive.google.com/uc?id=1lNHAz5QEo-OoHL8uAUuBPki-8xrEArhs) |\n| Ours (using `vgg16_from_caffe.npz`) | 91.0883 | 77.3528 | 65.3433 | 84.4276 | - |\n\n\u003cimg src=\"examples/voc/.readme/fcn32s_iter00092000.jpg\" width=\"30%\" /\u003e \u003cimg src=\"examples/voc/.readme/fcn16s_iter00100000.jpg\" width=\"30%\" /\u003e \u003cimg src=\"examples/voc/.readme/fcn8s_iter00072000.jpg\" width=\"30%\" /\u003e\n\nLeft to right, **FCN32s**, **FCN16s** and **FCN8s**, which are fully trained using this repo. See above tables to see the accuracy.\n\n\nLicense\n-------\n\nSee [LICENSE](LICENSE).\n\n\n## Cite This Project\n\nIf you use this project in your research or wish to refer to the baseline results published in the README, please use the following BibTeX entry.\n\n```bash\n@misc{chainer-fcn2016,\n  author =       {Ketaro Wada},\n  title =        {{fcn: Chainer Implementation of Fully Convolutional Networks}},\n  howpublished = {\\url{https://github.com/wkentaro/fcn}},\n  year =         {2016}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fwkentaro%2Ffcn","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fwkentaro%2Ffcn","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fwkentaro%2Ffcn/lists"}