{"id":32647547,"url":"https://github.com/jacobmarks/fiftyone_florence2_plugin","last_synced_at":"2025-10-31T05:55:13.533Z","repository":{"id":284241414,"uuid":"821900553","full_name":"jacobmarks/fiftyone_florence2_plugin","owner":"jacobmarks","description":"Run SOTA Vision-Language Model Florence-2 on your data!","archived":false,"fork":false,"pushed_at":"2025-03-24T21:39:16.000Z","size":6,"stargazers_count":9,"open_issues_count":0,"forks_count":1,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-03-24T22:29:13.259Z","etag":null,"topics":["computer-vision","datacentric","fiftyone-datasets","florence-2","ml","transformer","vision-language-model"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","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/jacobmarks.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}},"created_at":"2024-06-29T19:01:46.000Z","updated_at":"2025-03-24T21:39:22.000Z","dependencies_parsed_at":"2025-03-24T22:39:17.268Z","dependency_job_id":null,"html_url":"https://github.com/jacobmarks/fiftyone_florence2_plugin","commit_stats":null,"previous_names":["jacobmarks/fiftyone_florence2_plugin"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/jacobmarks/fiftyone_florence2_plugin","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jacobmarks%2Ffiftyone_florence2_plugin","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jacobmarks%2Ffiftyone_florence2_plugin/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jacobmarks%2Ffiftyone_florence2_plugin/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jacobmarks%2Ffiftyone_florence2_plugin/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/jacobmarks","download_url":"https://codeload.github.com/jacobmarks/fiftyone_florence2_plugin/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/jacobmarks%2Ffiftyone_florence2_plugin/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":281937758,"owners_count":26586774,"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","status":"online","status_checked_at":"2025-10-31T02:00:07.401Z","response_time":57,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"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":["computer-vision","datacentric","fiftyone-datasets","florence-2","ml","transformer","vision-language-model"],"created_at":"2025-10-31T05:55:09.002Z","updated_at":"2025-10-31T05:55:13.520Z","avatar_url":"https://github.com/jacobmarks.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"## 🏛️ Florence2 Plugin\n\n![florence2_plugin_demo](assets/florence2-plugin.gif)\n\n### Plugin Overview\n\nThis plugin allows you to apply Florence2 directly to your FiftyOne datasets.\n\n## Installation\n\nIf you haven't already, install FiftyOne:\n\n```shell\npip install -U fiftyone transformers\u003c=4.49 accelerate einops timm torch \n```\n\nThen, install the plugin:\n\n```shell\nfiftyone plugins download https://github.com/jacobmarks/fiftyone_florence2_plugin\n```\n\nYou can also install requirements via:\n\n```shell\nfiftyone plugins requirements @jacobmarks/florence2 --install\n```\n\nThe Florence2 plugin integrates Microsoft's Florence2 Vision-Language Model with FiftyOne datasets, offering several powerful computer vision capabilities:\n\n1. **Caption Generation** (`CaptionWithFlorence2`)\n   - Generates image captions with different levels of detail\n   - Customizable output field names\n   - Supports both immediate and delegated execution\n\n2. **Object Detection** (`DetectWithFlorence2`)\n   - Multiple detection modes:\n     - Standard object detection with default classes\n     - Dense region captioning\n     - Open vocabulary detection with custom text prompts\n     - Region proposals\n   - Flexible output field configuration\n\n3. **OCR** (`OCRWithFlorence2`)\n   - Performs text detection in images\n   - Option to store region information (bounding boxes)\n   - Configurable output fields\n\n4. **Caption-to-Phrase Grounding** (`CaptionToPhraseGroundingWithFlorence2`)\n   - Grounds specific phrases or captions by detecting relevant objects\n   - Supports both direct text input and existing string fields from the dataset\n\n5. **Referring Expression Segmentation** (`ReferringExpressionSegmentationWithFlorence2`)\n   - Performs segmentation based on textual descriptions\n   - Accepts either direct text input or references to existing dataset fields\n   - Outputs segmentation masks for the described objects\n\nAll operators support:\n- Custom model paths (defaults to \"microsoft/Florence-2-base-ft\")\n- Delegated execution for resource-intensive tasks\n- Flexible output field naming\n- Integration with FiftyOne's dataset operations\n\n## ℹ👨🏽‍💻 Refer to the [examples notebook](example_sdk_operators.ipynb) for detailed examples for how to run each operator via the FiftyOne SDK!\n\n\n# Citation\n\n```bibtext\n@article{xiao2023florence,\n  title={Florence-2: Advancing a unified representation for a variety of vision tasks},\n  author={Xiao, Bin and Wu, Haiping and Xu, Weijian and Dai, Xiyang and Hu, Houdong and Lu, Yumao and Zeng, Michael and Liu, Ce and Yuan, Lu},\n  journal={arXiv preprint arXiv:2311.06242},\n  year={2023}\n}\n```","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjacobmarks%2Ffiftyone_florence2_plugin","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fjacobmarks%2Ffiftyone_florence2_plugin","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fjacobmarks%2Ffiftyone_florence2_plugin/lists"}