- "Real-time Yoga recognition using deep learning" by Santosh Kumar Yadav, Amitojdeep Singh, Abhishek Gupta & Jagdish Lal Raheja
- "Yoga-82: A New Dataset for Fine-grained Classification of Human Poses" by Manisha Verma, Sudhakar Kumawat, Yuta Nakashima, Shanmuganathan Raman
- Videos used can be found here: https://www.researchgate.net/deref/https%3A%2F%2Farchive.org%2Fdetails%2FYogaVidCollected
- Video frames used can be found here: https://drive.google.com/drive/folders/1S6r6YH8upulktnfDbEn5DRxwQgqeXZN3?usp=sharing
- Steps to install OpenPose on the system can be found here: https://github.com/CMU-Perceptual-Computing-Lab/openpose
- to get 25 joint data, model used should be: BODY_25
- to get 18 joint data, model used should be: COCO
- 6 classes, 15 participants
- Duration of each video: 45+ seconds with 30 frames per second
- Surroundings: Indoor environment
- Total videos: 88
- Combined time: 1 hour, 6 minutes, and 5 seconds at 30 fps
- Data split ratio: 60:20:20 (approx.)
Results of the CNN-LSTM model on the 18 joint data of OpenPose are given below:
- Train Accuracy: 99.95%
- Validation Accuracy: 99.60%
- Test Accuracy: 99.87%