{"id":13510926,"url":"https://github.com/neu-vi/ezflow","last_synced_at":"2025-03-30T19:30:36.096Z","repository":{"id":38120952,"uuid":"383816972","full_name":"neu-vi/ezflow","owner":"neu-vi","description":"A modular PyTorch library for optical flow estimation using neural networks","archived":false,"fork":false,"pushed_at":"2024-04-08T15:55:00.000Z","size":609,"stargazers_count":134,"open_issues_count":25,"forks_count":9,"subscribers_count":5,"default_branch":"main","last_synced_at":"2025-02-13T09:47:14.746Z","etag":null,"topics":["modular","optical-flow","pytorch"],"latest_commit_sha":null,"homepage":"https://ezflow.readthedocs.io/","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/neu-vi.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.rst","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":"CITATION.cff","codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":"AUTHORS.rst","dei":null,"publiccode":null,"codemeta":null}},"created_at":"2021-07-07T13:58:07.000Z","updated_at":"2025-01-05T12:43:37.000Z","dependencies_parsed_at":"2023-01-31T09:16:06.907Z","dependency_job_id":"5599145b-0b91-4c6d-9ca9-8569d2430747","html_url":"https://github.com/neu-vi/ezflow","commit_stats":{"total_commits":252,"total_committers":6,"mean_commits":42.0,"dds":"0.32539682539682535","last_synced_commit":"dfceac14170d1e46cfcb775257a5769295480db2"},"previous_names":["neu-vig/ezflow"],"tags_count":6,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/neu-vi%2Fezflow","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/neu-vi%2Fezflow/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/neu-vi%2Fezflow/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/neu-vi%2Fezflow/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/neu-vi","download_url":"https://codeload.github.com/neu-vi/ezflow/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246368604,"owners_count":20766053,"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":["modular","optical-flow","pytorch"],"created_at":"2024-08-01T03:00:22.330Z","updated_at":"2025-03-30T19:30:35.737Z","avatar_url":"https://github.com/neu-vi.png","language":"Python","funding_links":[],"categories":["Papers \u0026 Code"],"sub_categories":["Optical flow toolkit"],"readme":"\u003cp align=\"center\"\u003e\n    \u003cbr\u003e\n    \u003cimg src=\"./docs/assets/logo.png\" height=\"60\" width=\"60\"/\u003e\n    \u003cbr\u003e\n\u003c/p\u003e\n\n\u003ch1 align=\"center\"\u003eEzFlow\u003c/h1\u003e\n\u003ch3 align=\"center\"\u003eA modular PyTorch library for optical flow estimation using neural networks\u003c/h3\u003e\n\n\u003cdiv align='center'\u003e\n\n[![Tests](https://github.com/neu-vig/ezflow/actions/workflows/package-test.yml/badge.svg)](https://github.com/neu-vig/ezflow/actions/workflows/package-test.yml)\n[![Docs](https://readthedocs.org/projects/ezflow/badge/?version=latest)](https://ezflow.readthedocs.io/en/latest/?badge=latest)\n[![Downloads](https://static.pepy.tech/badge/ezflow)](https://pepy.tech/project/ezflow)  \n\n\u003c!-- [![Code style](https://github.com/neu-vig/ezflow/actions/workflows/linting.yml/badge.svg)](https://github.com/neu-vig/ezflow/actions/workflows/linting.yml) --\u003e\n\u003c!-- [![Code coverage](https://github.com/neu-vig/ezflow/actions/workflows/codecov.yml/badge.svg)](https://github.com/neu-vig/ezflow/actions/workflows/codecov.yml) --\u003e\n\n**[Documentation](https://ezflow.readthedocs.io/en/latest/)** | **[Tutorials](https://ezflow.readthedocs.io/en/latest/tutorials/index.html)**\n\n\u003c/div\u003e\n\n\n## Installation\n\n### From source (recommended)\n\n```shell\n\ngit clone https://github.com/neu-vig/ezflow\ncd ezflow/\npython setup.py install\n\n```\n\n### From PyPI\n\n```shell\n\npip install ezflow\n\n```\n___\n\n### Models supported\n\n- [x] [DICL](https://arxiv.org/abs/2010.14851)\n- [x] [DCVNet](https://jianghz.me/files/DCVNet_camera_ready_wacv2023.pdf) ([1 checkpoint](./configs/README.md))\n- [x] [FlowNetS](https://arxiv.org/abs/1504.06852)\n- [x] [FlowNetC](https://arxiv.org/abs/1504.06852) ([3 checkpoints](./configs/README.md))\n- [x] [PWCNet](https://arxiv.org/abs/1709.02371) ([3 checkpoints](./configs/README.md)) \n- [x] [RAFT](https://arxiv.org/abs/2003.12039) ([3 checkpoints](./configs/README.md))\n- [x] [VCN](https://papers.nips.cc/paper/2019/hash/bbf94b34eb32268ada57a3be5062fe7d-Abstract.html)\n\n### Datasets supported\n\n- [x] [AutoFlow](https://autoflow-google.github.io/)\n- [x] [FlyingChairs](https://lmb.informatik.uni-freiburg.de/resources/datasets/FlyingChairs.en.html#flyingchairs)\n- [x] [HD1K](http://hci-benchmark.iwr.uni-heidelberg.de/)\n- [x] [KITTI](http://www.cvlibs.net/datasets/kitti/eval_scene_flow.php?benchmark=flow)\n- [x] [Kubric](https://github.com/google-research/kubric)\n- [x] [MPI Sintel](http://sintel.is.tue.mpg.de/)\n- [x] [SceneFlow Monkaa](https://lmb.informatik.uni-freiburg.de/resources/datasets/SceneFlowDatasets.en.html)\n- [x] [SceneFlow Driving](https://lmb.informatik.uni-freiburg.de/resources/datasets/SceneFlowDatasets.en.html)\n- [x] [SceneFlow FlyingThings3D](https://lmb.informatik.uni-freiburg.de/resources/datasets/SceneFlowDatasets.en.html)\n- [x] [SceneFlow FlyingThings3D subset](https://lmb.informatik.uni-freiburg.de/resources/datasets/SceneFlowDatasets.en.html)\n\n___\n\n### Results and Pre-trained checkpoints\n\n- #### DCVNet | [model config](./configs/models/dcvnet.yaml) | [paper](https://jianghz.me/files/DCVNet_camera_ready_wacv2023.pdf)\n| Training Dataset                        | Training Config                                                         | ckpts                                                                                  | Sintel Clean (training) | Sintel Final(training)| KITTI2015 AEPE | KITTI2015 F1-all |\n|-----------------------------------------|-------------------------------------------------------------------------|----------------------------------------------------------------------------------------|-------------------------|-----------------------|----------------|------------------|\n| FlyingThings3DSubset + Monkaa + Driving | [config](./configs/trainers/dcvnet/dcvnet_sceneflow_baseline.yaml)      | [download](https://jianghz.me/files/ezflow_ckpts/dcvnet_sceneflow_step800k.pth)        | 1.90                    | 3.35                  | 4.75           | 23.41%           |\n\n- #### FlowNetC | [model config](./configs/models/flownet_c.yaml) | [arXiv](https://arxiv.org/abs/1504.06852)\n\n| Training Dataset | Training Config                                                         | ckpts                                                                                  | Sintel Clean (training) | Sintel Final(training)| KITTI2015 AEPE | KITTI2015 F1-all |\n|------------------|-------------------------------------------------------------------------|----------------------------------------------------------------------------------------|-------------------------|-----------------------|----------------|------------------|\n| Chairs           | [config](./configs/trainers/flownetc/flownetc_chairs_baseline.yaml)     | [download](https://jianghz.me/files/ezflow_ckpts/flownetc_chairs_step1200k.pth)        | 3.41                    | 4.94                  | 14.84          | 54.23%           |\n| Chairs -\u003e Things | [config](./configs/trainers/flownetc/flownetc_things_baseline.yaml)     | [download](https://jianghz.me/files/ezflow_ckpts/flownetc_chairs_things_step1574k.pth) | 2.93                    | 4.48                  | 12.47          | 45.89%           |\n| Kubric           | [config](./configs/trainers/flownetc/flownetc_kubric_improved_aug.yaml) | [download](https://jianghz.me/files/ezflow_ckpts/flownetc_kubric_step1200k.pth)        | 3.57                    | 3.96                  | 12.11          | 36.35%           |\n\n- #### PWC-Net | [model config](./configs/models/pwcnet.yaml)  | [arXiv](https://arxiv.org/abs/1709.02371)\n\n| Training Dataset | Training Config                                                     | ckpts                                                                               | Sintel Clean (training) | Sintel Final(training)| KITTI2015 AEPE | KITTI2015 F1-all |\n|------------------|---------------------------------------------------------------------|-------------------------------------------------------------------------------------|-------------------------|-----------------------|----------------|------------------|\n| Chairs           | [config](./configs/trainers/pwcnet/pwcnet_chairs_baseline.yaml)     | [download](https://jianghz.me/files/ezflow_ckpts/pwcnet_chairs_step1200k.pth)       | 3.5                     | 4.73                  | 17.81          | 51.76%           |\n| Chairs -\u003e Things | [config](./configs/trainers/pwcnet/pwcnet_things_baseline.yaml)     | [download](https://jianghz.me/files/ezflow_ckpts/pwcnet_chairs_things_step2400k.pth)| 2.06                    | 3.43                  | 11.04          | 32.68%           |\n| Kubric           | [config](./configs/trainers/pwcnet/pwcnet_kubric_improved_aug.yaml) | [download](https://jianghz.me/files/ezflow_ckpts/pwcnet_kubric_step1200k.pth)       | 3.08                    | 3.31                  | 9.83           | 21.94%           |\n\n\n- #### RAFT | [model config](./configs/models/raft.yaml) | [arXiv](https://arxiv.org/abs/2003.12039)\n\n| Training Dataset | Training Config                                                 | ckpts                                                                                | Sintel Clean (training) | Sintel Final(training)| KITTI2015 AEPE | KITTI2015 F1-all |\n|------------------|-----------------------------------------------------------------|--------------------------------------------------------------------------------------|-------------------------|-----------------------|----------------|------------------|\n| Chairs           | [config](./configs/trainers/raft/raft_chairs_baseline.yaml)     | [download](https://jianghz.me/files/ezflow_ckpts/raft_chairs_step100k_v2.pth)        | 2.23                    | 4.56                  | 10.45          | 38.93%           |\n| Chairs -\u003e Things | [config](./configs/trainers/raft/raft_things_baseline.yaml)     | [download](https://jianghz.me/files/ezflow_ckpts/raft_chairs_things_step200k_v2.pth) | 1.66                    | 2.75                  | 5.01           | 16.87%           |\n| Kubric           | [config](./configs/trainers/raft/raft_kubric_improved_aug.yaml) | [download](https://jianghz.me/files/ezflow_ckpts/raft_kubric_step100k_v2.pth)        | 2.12                    | 2.54                  | 6.01           | 17.35%           |\n\n___\n\n#### Additional Information\n\n- KITTI dataset has been evaluated with a center crop of size `1224 x 370`.\n- FlowNetC and PWC-Net uses `padding` of size `64` for evaluating the KITTI2015 dataset.\n- RAFT and DCVNet uses `padding` of size `8` for evaluating the Sintel and KITTI2015 datasets.\n___\n### References\n\n- [RAFT](https://github.com/princeton-vl/RAFT)\n- [DICL-Flow](https://github.com/jytime/DICL-Flow)\n- [PWC-Net](https://github.com/NVlabs/PWC-Net)\n- [FlowNetPytorch](https://github.com/ClementPinard/FlowNetPytorch)\n- [VCN](https://github.com/gengshan-y/VCN)\n- [detectron2](https://github.com/facebookresearch/detectron2)\n- [CorrelationLayer](https://github.com/oblime/CorrelationLayer)\n- [ptflow](https://github.com/hmorimitsu/ptlflow)\n\n\n\n\n\u003cbr\u003e\n\n\u003cfooter\u003e\n\u003ca target=\"_blank\" href=\"https://icons8.com/icon/3Nj3FNnz36Id/pixels\"\u003ePixels\u003c/a\u003e icon by \u003ca target=\"_blank\" href=\"https://icons8.com\"\u003eIcons8\u003c/a\u003e\n\u003c/footer\u003e\n\n\n\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fneu-vi%2Fezflow","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fneu-vi%2Fezflow","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fneu-vi%2Fezflow/lists"}