This code helps you to get hand landmark especially when playing piano or violin.
- Python
- OpenCV
- Mediapipe
- FFmpeg
video_to_csv performs hand shape recognition using the MediaPipe Hands library.
It recognizes hand shapes for a given set of images, extracts landmark locations for each hand, and stores them as data.
audio extracts audio from a video file using FFmpeg.
It extracts 1 second of audio from a given video file and saves the extracted audio as an MP3 file.
MFCC saves the MFCC features extracted from the file, reads skeleton data from a CSV file, and extracts data for the desired columns from a specified range of frames.
