{"id":13724404,"url":"https://github.com/pedropro/TACO","last_synced_at":"2025-05-07T18:31:46.512Z","repository":{"id":47125949,"uuid":"190940467","full_name":"pedropro/TACO","owner":"pedropro","description":"🌮 Trash Annotations in Context Dataset Toolkit","archived":false,"fork":false,"pushed_at":"2024-06-16T18:57:16.000Z","size":102753,"stargazers_count":641,"open_issues_count":39,"forks_count":212,"subscribers_count":26,"default_branch":"master","last_synced_at":"2025-05-04T16:02:09.617Z","etag":null,"topics":["dataset","deep-learning","garbage","litter","mask-rcnn","object-detection","trash"],"latest_commit_sha":null,"homepage":"http://tacodataset.org","language":"Jupyter Notebook","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/pedropro.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,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2019-06-08T22:19:04.000Z","updated_at":"2025-05-02T05:02:17.000Z","dependencies_parsed_at":"2024-11-05T23:10:56.743Z","dependency_job_id":null,"html_url":"https://github.com/pedropro/TACO","commit_stats":{"total_commits":185,"total_committers":3,"mean_commits":"61.666666666666664","dds":"0.021621621621621623","last_synced_commit":"29de1a9ba05a647b83a90f18d7772e20bb23d846"},"previous_names":[],"tags_count":2,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pedropro%2FTACO","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pedropro%2FTACO/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pedropro%2FTACO/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/pedropro%2FTACO/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/pedropro","download_url":"https://codeload.github.com/pedropro/TACO/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":252533843,"owners_count":21763664,"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":["dataset","deep-learning","garbage","litter","mask-rcnn","object-detection","trash"],"created_at":"2024-08-03T01:01:56.596Z","updated_at":"2025-05-07T18:31:46.478Z","avatar_url":"https://github.com/pedropro.png","language":"Jupyter Notebook","funding_links":[],"categories":["Circular Economy and Waste","Jupyter Notebook","Industrial Ecology"],"sub_categories":["Soil","Circular Economy and Waste"],"readme":"\u003cp align=\"center\"\u003e\n\u003cimg src=\"https://raw.githubusercontent.com/wiki/pedropro/TACO/images/logonav.png\" width=\"25%\"/\u003e\n\u003c/p\u003e\n\nTACO is a growing image dataset of waste in the wild. It contains images of litter taken under\ndiverse environments: woods, roads and beaches. These images are manually labeled and segmented\naccording to a hierarchical taxonomy to train and evaluate object detection algorithms. Currently,\nimages are hosted on Flickr and we have a server that is collecting more images and\nannotations @ [tacodataset.org](http://tacodataset.org)\n\n\n\u003cdiv align=\"center\"\u003e\n  \u003cdiv class=\"column\"\u003e\n    \u003cimg src=\"https://raw.githubusercontent.com/wiki/pedropro/TACO/images/1.png\" width=\"17%\" hspace=\"3\"\u003e\n    \u003cimg src=\"https://raw.githubusercontent.com/wiki/pedropro/TACO/images/2.png\" width=\"17%\" hspace=\"3\"\u003e\n    \u003cimg src=\"https://raw.githubusercontent.com/wiki/pedropro/TACO/images/3.png\" width=\"17%\" hspace=\"3\"\u003e\n    \u003cimg src=\"https://raw.githubusercontent.com/wiki/pedropro/TACO/images/4.png\" width=\"17%\" hspace=\"3\"\u003e\n    \u003cimg src=\"https://raw.githubusercontent.com/wiki/pedropro/TACO/images/5.png\" width=\"17%\" hspace=\"3\"\u003e\n  \u003c/div\u003e\n\u003c/div\u003e\n\u003c/br\u003e\n\nFor convenience, annotations are provided in COCO format. Check the metadata here:\nhttp://cocodataset.org/#format-data\n\nTACO is still relatively small, but it is growing. Stay tuned!\n\n# Publications\n\nFor more details check our paper: https://arxiv.org/abs/2003.06975\n\nIf you use this dataset and API in a publication, please cite us using: \u0026nbsp;\n```\n@article{taco2020,\n    title={TACO: Trash Annotations in Context for Litter Detection},\n    author={Pedro F Proença and Pedro Simões},\n    journal={arXiv preprint arXiv:2003.06975},\n    year={2020}\n}\n```\n\n# News\n**December 20, 2019** - Added more 785 images and 2642 litter segmentations. \u003cbr/\u003e\n**November 20, 2019** - TACO is officially open for new annotations: http://tacodataset.org/annotate\n\n# Getting started\n\n### Requirements \n\nTo install the required python packages simply type\n```\npip3 install -r requirements.txt\n```\nAdditionaly, to use ``demo.pynb``, you will also need [coco python api](https://github.com/cocodataset/cocoapi). You can get this using\n```\npip3 install git+https://github.com/philferriere/cocoapi.git#subdirectory=PythonAPI\n```\n\n### Download\n\nTo download the dataset images simply issue\n```\npython3 download.py\n```\nAlternatively, download from [![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.3587843.svg)](https://doi.org/10.5281/zenodo.3587843)\n\nOur API contains a jupyter notebook ``demo.pynb`` to inspect the dataset and visualize annotations.\n\n**Unlabeled data**\n\nA list of URLs for both unlabeled and labeled images is now also provided in `data/all_image_urls.csv`.\nEach image contains one URL for each original image (second column) and one URL for a VGA-resized version (first column)\nfor images hosted by Flickr. If you decide to annotate these images using other tools, please make them public and contact us so we can keep track.\n\n**Unofficial data**\n\nAnnotations submitted via our website are added weekly to `data/annotations_unofficial.json`. These have not yet been been reviewed by us -- some may be inaccurate or have poor segmentations. \nYou can use the same command to download the respective images:\n```\npython3 download.py --dataset_path ./data/annotations_unofficial.json\n```\n\n### Trash Detection\n\nThe implementation of [Mask R-CNN by Matterport](https://github.com/matterport/Mask_RCNN)  is included in ``/detector``\nwith a few modifications. Requirements are the same. Before using this, the dataset needs to be split. You can either donwload our [weights and splits](https://github.com/pedropro/TACO/releases/tag/1.0) or generate these from scratch using the `split_dataset.py` script to generate \nN random train, val, test subsets. For example, run this inside the directory `detector`:\n```\npython3 split_dataset.py --dataset_dir ../data\n```\n\nFor further usage instructions, check ``detector/detector.py``.\n\nAs you can see [here](http://tacodataset.org/stats), most of the original classes of TACO have very few annotations, therefore these must be either left out or merged together. Depending on the problem, ``detector/taco_config`` contains several class maps to target classes, which maintain the most dominant classes, e.g., Can, Bottles and Plastic bags. Feel free to make your own classes.\n\n\u003cp align=\"center\"\u003e\n\u003cimg src=\"https://raw.githubusercontent.com/wiki/pedropro/TACO/images/teaser.gif\" width=\"75%\"/\u003e\u003c/p\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpedropro%2FTACO","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fpedropro%2FTACO","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fpedropro%2FTACO/lists"}