This project develops a low-cost acoustic camera that combines multi-channel audio sensing with computer vision to localize and interpret sound sources in real time. By fusing beamforming-based acoustic heatmaps with machine-learning-driven object and sound classification, the system identifies what is producing sound and where it is located within a scene.
Team S03 Members: Wendong Wang, Xiaoyan Li, Zihan Wang, Lerong Qi, Zuyu Fu
Capstone Advisor: Shahram Shirani