Scene understanding based on visual and acoustic data

• Main Author: Feng, Nijing
• Other Authors: Mao Kezhi
• Format: Final Year Project
• Language: English
• Published: 2019
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/77672
• Institution: Nanyang Technological University

Convolutional Neural Networks (CNN) is the latest development of neural network. It is a deep learning network having multiple layers of convolution operations to extract image or video features and make predictions. With recent years of rapid development, CNN can be seen everywhere with various applications in real life. New models of CNNs are being developed continuously. In this report, the basic principles of CNN will be discussed, experiments performed will be described, attempts to improve CNN performance with additional acoustic data will be explained.