Skip to content

mansikataria/tongue-tip-detection-and-tracking

Repository files navigation

Tongue Tip Detection and Tracking

This project is for detection of tip of tongue in real time and video, and track it. Libraries used in this project are: Python, OpenCV, dlib, numpy, matplotlib

Real time Demo:

How to run?

For Real time detection:

  1. The default camera used is in-build webcam (0), if you re using usb webcam (1) set the camera number on line 12 in detect-tongue-tip-real-time.py
  2. Execute python detect-tongue-tip-real-time.py; press q to escape anytime

For existing video

  1. Get a video you want to input.
  2. Change line 12 in detect-tongue-with-blob-tracking-revised.py to use your video
  3. Just execute python3 detect-tongue-with-blob-tracking-revised.py

Overview of the logic:

  1. Divide the video into frames
  2. Fetch mouth part
  3. Detect blob keypoints in the mouth part using OpenCV
  4. Based on detections in step 2 and 3, an algorithm is implemented in python, numpy around the height of inner mouth, width of mouth, blob keypoint to detect the tip of tongue