this project is using opencv & mediapipe_holistic model to access webcam and capture people action including face, pose and hands.
the model is mainly built by LSTM layers to recognize actions which includes some NBA player's iconic celebration moves.
- Finger wag 0
- The Shrug
- Three(Three to the dome)
- Three goggles
- Home cooking