{"id":13778144,"url":"https://github.com/chunbolang/HPA","last_synced_at":"2025-05-11T11:35:00.953Z","repository":{"id":48146140,"uuid":"402261977","full_name":"chunbolang/HPA","owner":"chunbolang","description":"Official PyTorch Implementation of Holistic Prototype Activation for Few-Shot Segmentation (TPAMI'22).","archived":false,"fork":false,"pushed_at":"2022-07-22T07:17:50.000Z","size":8200,"stargazers_count":20,"open_issues_count":3,"forks_count":3,"subscribers_count":1,"default_branch":"main","last_synced_at":"2024-11-17T13:38:14.125Z","etag":null,"topics":["computer-vision","few-shot-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/chunbolang.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":"2021-09-02T02:15:42.000Z","updated_at":"2024-09-02T12:51:35.000Z","dependencies_parsed_at":"2022-08-26T09:10:41.887Z","dependency_job_id":null,"html_url":"https://github.com/chunbolang/HPA","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/chunbolang%2FHPA","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chunbolang%2FHPA/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chunbolang%2FHPA/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chunbolang%2FHPA/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/chunbolang","download_url":"https://codeload.github.com/chunbolang/HPA/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":253559921,"owners_count":21927701,"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":["computer-vision","few-shot-segmentation"],"created_at":"2024-08-03T18:00:51.530Z","updated_at":"2025-05-11T11:35:00.060Z","avatar_url":"https://github.com/chunbolang.png","language":"Python","funding_links":[],"categories":["2022"],"sub_categories":[],"readme":"# Holistic Prototype Activation for Few-Shot Segmentation\n\nThis repo contains the code for our **IEEE TPAMI 2022** paper \"*Holistic Prototype Activation for Few-Shot Segmentation*\" by Gong Cheng, Chunbo Lang, and Junwei Han.\n\n## 📋 Note\n\nPlease refer to our BAM [repository](https://github.com/chunbolang/BAM) for the latest **training/testing** scripts. HPA can also be naturally combined with BAM (state-of-the-art) as a stronger meta-learner, with potential for further improvement.\n\n##\n### Dependencies\n\n- Python 3.6\n- PyTorch 1.3.1\n- cuda 9.0\n- torchvision 0.4.2\n- tensorboardX 2.1\n\n### Datasets\n\n- PASCAL-5i:  [VOC2012](http://host.robots.ox.ac.uk/pascal/VOC/voc2012/) + [SBD](http://home.bharathh.info/pubs/codes/SBD/download.html)\n- COCO-20i:  [COCO2014](https://cocodataset.org/#download)\n\n   Please see [OSLSM](https://arxiv.org/abs/1709.03410) and [FWB](https://openaccess.thecvf.com/content_ICCV_2019/html/Nguyen_Feature_Weighting_and_Boosting_for_Few-Shot_Segmentation_ICCV_2019_paper.html) for more details on datasets. \n\n### Usage\n\n1. Download the prior prototypes of base categories from our [Google Drive](https://drive.google.com/file/d/11-VHCAAO6NcnP2OzZdT2rNrGpC9LqKPh/view?usp=sharing) and put them under `HPA/initmodel/prototypes`. \n2. Download the pre-trained backbones from [here](https://drive.google.com/file/d/1AQcvMHHpURZM67MMgV-S3T0Kz-h2q7FR/view?usp=sharing).\n3. Change configuration via the `.yaml` files in `HPA/config`, then run the `.sh` scripts for training and testing.\n\n### To-Do List\n\n- [x] Support different backbones\n- [x] Support various annotations for training/testing\n- [ ] Zero-Shot Segmentation (ZSS)\n- [ ] FSS-1000 dataset\n- [ ] Multi-GPU training\n\n### References\n\nThis repo is built based on [PFENet](https://github.com/dvlab-research/PFENet) and [DANet](https://github.com/junfu1115/DANet). Thanks for their great work!\n\n### BibTeX\n\nIf you find our work and this repository useful. Please consider giving a star :star: and citation \u0026#x1F4DA;.\n\n```bibtex\n@article{lang2022hpa,\n  title={Holistic Prototype Activation for Few-Shot Segmentation},\n  author={Cheng, Gong and Lang, Chunbo and Han, Junwei},\n  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},\n  year={2022},\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fchunbolang%2FHPA","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fchunbolang%2FHPA","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fchunbolang%2FHPA/lists"}