Yawn Detection Using Support Vector Machine Classification
Abstract
A model to detect yawning in a video sequence from a regular camera is proposed. Landmark detectors are able to quickly and accurately estimate the pose of a face with various amounts of light and facial expressions. We can show that the information from these landmark detectors is useful in being able to detect when a person is yawning. The proposed model is therefore able to take in information from the landmark detector and process it by creating new features to characterize the yawning in each frame. A support vector machine classifier then detects yawning as a pattern of the constructed feature values in a short temporal window. We made a dataset and annotated it to train and test the model, which is composed of frames from several videos that include a person yawning and not yawning at different angles to the camera.
Subject
YawnYawning
Yawn Detection
Computer Vision
Machine Learning
Deep Learning
Artificial Intelligence
Data Science
Computer Science
Support Vector Machine
SVM
AI
Citation
Mayilvahanan, Sarvesh (2022). Yawn Detection Using Support Vector Machine Classification. Undergraduate Research Scholars Program. Available electronically from https : / /hdl .handle .net /1969 .1 /188407.