CLASSIFICATION OF EMOTIONS FROM LIVE VIDEO USING MACHINE LEARNING
Karri Rekha Sai Ratnam, Chiraparapu Srinivasa Rao · Journal of Emerging Technologies and Innovative Research · 2021
Visual sentiment analysis, which investigates humans' emotional responses to visual stimuli such as images and videos, has been a fascinating and challenging problem. It attempts to recognize the high-level content of visual data. The success of current models can be attributed to the development of robust computer vision algorithms. The majority of existing models attempt to solve the problem by recommending either robust features or more complex models. The main proposed inputs are visual features from the entire image or video. Local areas have received little attention, which we believe is important to the emotional response of humans to the entire image. Image recognition is used to find people in images and analyse their sentiments or emotions. The CNN algorithm is used in this project to accomplish this task. Given an image, it will search for faces, identify them, place a rectangle in their positions, and describe the emotion found with a percentage of emotions displayed.