The purpose of this project is to develop web-based demonstration to visualize intuitive and fluid control over 3D object using hand-based gesture controls. Project focuses on collection of data to train model, development of gesture detection model using Keras, and deployment of said model in a web app. Gesture classification is done via sequential dense neural network with three hidden layers; whereas its deployment in the form of web app is done using React JS and Three JS. End goal of project is to create a web app which can control a 3D object using hand gesture fed from webcam.
Model Repo
yarn install
yarn start
⚠️ **If not working please make sure web cam access is given (as this is required)
- D. P. Kingma and B. Jimmy, " arXiv.org," Jan 2017. [Online]. Available: https://arxiv.org/abs/1412.6980. [Accessed 03 Jan 2022].
- M. Patrick, "mediapipe," Google LLC, [Online]. Available: https://google.github.io/mediapipe/solutions/hands. [Accessed 03 Jan 2022].
- C. and F. , "Tensorflow," 2015. [Online]. Available: https://github.com/fchollet/keras. [Accessed 03 Jan 2022].