{"id":18347376,"url":"https://github.com/marc-kruiss/signlanguage-actiondetection","last_synced_at":"2026-05-08T14:40:16.645Z","repository":{"id":154362256,"uuid":"512766117","full_name":"Marc-Kruiss/SignLanguage-ActionDetection","owner":"Marc-Kruiss","description":"This project allows to train and test sign language data to identify numbers, the alphabet and poses with the help of opencv and mediapipe","archived":false,"fork":false,"pushed_at":"2023-04-03T17:19:48.000Z","size":6598,"stargazers_count":2,"open_issues_count":0,"forks_count":0,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-04-09T23:38:23.895Z","etag":null,"topics":["action-recognition","ai","artificial-intelligence","lstm","lstm-neural-networks","mediapipe","mediapipe-hands","opencv","python"],"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/Marc-Kruiss.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":"2022-07-11T13:23:10.000Z","updated_at":"2025-01-11T05:45:12.000Z","dependencies_parsed_at":null,"dependency_job_id":"866b88c8-0901-4cdb-bfdb-2e62f6885d44","html_url":"https://github.com/Marc-Kruiss/SignLanguage-ActionDetection","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/Marc-Kruiss/SignLanguage-ActionDetection","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Marc-Kruiss%2FSignLanguage-ActionDetection","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Marc-Kruiss%2FSignLanguage-ActionDetection/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Marc-Kruiss%2FSignLanguage-ActionDetection/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Marc-Kruiss%2FSignLanguage-ActionDetection/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Marc-Kruiss","download_url":"https://codeload.github.com/Marc-Kruiss/SignLanguage-ActionDetection/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Marc-Kruiss%2FSignLanguage-ActionDetection/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":264740458,"owners_count":23656778,"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":["action-recognition","ai","artificial-intelligence","lstm","lstm-neural-networks","mediapipe","mediapipe-hands","opencv","python"],"created_at":"2024-11-05T21:13:51.507Z","updated_at":"2026-05-08T14:40:11.623Z","avatar_url":"https://github.com/Marc-Kruiss.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Real-time Sign Language Detection\nThis project is written in Python and designed to detect sign language \ngestures in real-time through a webcam video stream. It uses a pre-trained \nkeras model to identify three gestures: \"hello\", \"I love you\", and \"thanks\". \nThe detected gestures are displayed on the screen, allowing the user to build \nsentences with them.\n\n## Packages\nThe program uses the following packages:\n\n* **OpenCV**: OpenCV is a popular computer vision library that is used for real-time computer vision applications. It is used in this project for webcam feed and information display.\n\n* **MediaPipe**: MediaPipe is an open-source framework that provides cross-platform, customizable ML solutions for live and streaming media. It is used in this project for gesture keypoints detection.\n\n* **Keras**: Keras is a high-level neural networks API, written in Python and capable of running on top of TensorFlow, CNTK, or Theano. It is used in this project for machine learning.\n\n## Usage\nThere are four main scripts in the program:\n\n* `make_predictions.py`: Predicts the trained gestures and builds sentences.\n* `folder_setup.py`: Builds the folder structure.\n* `training_testing.py`: Trains the model with the \"MP_Data\" Keypoints and evaluates it.\n* `setup_training_testing_keypoints.py`: Starts a training program where the user can input the gestures for the actions. The keypoints are then saved and can be used for training and evaluation.\n\n## Getting started\nTo get started with the program, follow these steps:\n\n1. Install the required packages using the following command:\u003cbr\u003e\n```pip install opencv-python mediapipe keras```\n\n2. Clone the repository and navigate to the project directory.\u003cbr\u003e\n`git clone https://github.com/Marc-Kruiss/SignLanguage-ActionDetection`\u003cbr\u003e\n```cd SignLanguage-ActionDetection```\n\n3. Run the `folder_setup` script to build the required folder structure.\u003cbr\u003e\n```python folder_setup.py```\n\n4. Run the `setup_training_testing_keypoints` script to input the gestures for the actions and save the keypoints.\u003cbr\u003e\n```python setup_training_testing_keypoints.py```\n\n5. Train the model using the saved keypoints by running the `training_testing.py` script.\u003cbr\u003e\n```python training_testing.py```\n\n6. Use the `make_predictions` script to detect the gestures in real-time and build sentences.\u003cbr\u003e\n```python make_predictions.py```\n\n## References\nMediaPipe: https://mediapipe.dev/ \u003cbr\u003e\nKeras: https://keras.io/ \u003cbr\u003e\nOpenCV: https://opencv.org/ \u003cbr\u003e\n\n## Acknowledgements\nThis project was inspired by the work of Ahmed Hassanien and his article \"Real-Time American Sign Language Recognition using Deep Learning Neural Networks\"","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmarc-kruiss%2Fsignlanguage-actiondetection","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fmarc-kruiss%2Fsignlanguage-actiondetection","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fmarc-kruiss%2Fsignlanguage-actiondetection/lists"}