{"id":27933519,"url":"https://github.com/thoth000/shape-aware-refinement","last_synced_at":"2026-05-19T09:11:48.768Z","repository":{"id":285179680,"uuid":"957305548","full_name":"thoth000/Shape-Aware-Refinement","owner":"thoth000","description":"[JSAI2025] Research code for shape-aware refinement in segmentation using PDEs.","archived":false,"fork":false,"pushed_at":"2025-05-17T17:46:48.000Z","size":3707,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-07-23T07:31:44.498Z","etag":null,"topics":["anisotropic-diffusion","pytorch","segmentation"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/thoth000.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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,"zenodo":null}},"created_at":"2025-03-30T03:05:48.000Z","updated_at":"2025-05-17T17:45:57.000Z","dependencies_parsed_at":"2025-05-07T04:59:10.357Z","dependency_job_id":"16afe971-c0d6-41a6-9e58-8ef7bce864a7","html_url":"https://github.com/thoth000/Shape-Aware-Refinement","commit_stats":null,"previous_names":["thoth000/shape-aware-refinement"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/thoth000/Shape-Aware-Refinement","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/thoth000%2FShape-Aware-Refinement","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/thoth000%2FShape-Aware-Refinement/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/thoth000%2FShape-Aware-Refinement/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/thoth000%2FShape-Aware-Refinement/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/thoth000","download_url":"https://codeload.github.com/thoth000/Shape-Aware-Refinement/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/thoth000%2FShape-Aware-Refinement/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":33209675,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-19T07:54:09.561Z","status":"ssl_error","status_checked_at":"2026-05-19T07:54:08.508Z","response_time":58,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"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":["anisotropic-diffusion","pytorch","segmentation"],"created_at":"2025-05-07T04:59:08.325Z","updated_at":"2026-05-19T09:11:48.749Z","avatar_url":"https://github.com/thoth000.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Shape-Aware Refinement\nThis research proposes a **PDE-based trainable refinement module** for **tubular-structure segmentation** to enhance shape continuity.\n![image](https://github.com/user-attachments/assets/3865efd2-732f-470a-9b96-36f107b2e5c5)\n\n## Paper / Slides\n- [Presentation Slides](https://github.com/thoth000/Shape-Aware-Refinement/blob/main/docs/presentation_ja.pdf)\n- [JSAI 2025 Abstract (Japanese)](https://github.com/thoth000/Shape-Aware-Refinement/blob/main/docs/jsai_abstract.md)\n\n## Overview\n- PDE-based refinement improves mask continuity **while maintaining thin shape**\n- Our approach is applied to **various architectures** for tubular-structure segmentation because of **post-process**\n\n## Dataset\n- [DRIVE dataset](https://github.com/zhengyuan-liu/Retinal-Vessel-Segmentation/tree/master/DRIVE)\n\n## Environment\n- Ubuntu 24.04.2 LTS\n- NVIDIA TITAN RTX(Memory: 24GB) * 4\n\n## How to run\n1. setup the environment.\n  ```bash\n  $ pip install -r requirements.txt\n  ``` \n\n2. Download DRIVE dataset to `/dataset`.\n\n3. Run `/dataset/data_process.py` to fit dataset for training models.\n  ```bash\n  $ cd dataset\n  $ python data_process.py -dp DATASET_PATH -dn DATASET_NAME\n  ```\n\n4. train and test model.\n  ```bash\n  $ cd pde-shape-refiner\n  $ main_ddp.sh\n  ```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fthoth000%2Fshape-aware-refinement","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fthoth000%2Fshape-aware-refinement","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fthoth000%2Fshape-aware-refinement/lists"}