Facial Keypoint Detection with Convolutional Neural Networks
Savina Jassica Colaco, Dong Seog Han · 2020
Facial keypoint detection is a challenging problem in the field of computer vision. The keypoint detection is done by predicting the coordinates of certain facial features. In this paper, facial keypoint detection is predicted using convolutional neural networks. The models are trained to predict facial key points using the webcam input data. The facial keypoints includes eyebrow corners, nose tip, eye corners and center, and lip points. The predicted keypoints are mapped onto the webcam input data and compared with different models for better detection of keypoints. The mean square error is used to estimate the loss of each model.