Justin Wilson

University of North Carolina at Chapel Hill

Papers

Total Citations

H-Index

About

Justin Wilson is a pioneering researcher at the intersection of robotics and multimodal perception, specializing in audio-visual learning for autonomous systems. His work fundamentally challenges the limitations of vision-only approaches by integrating sound as a complementary sensory modality. In his highly cited 2020 paper, "AVOT: Audio-Visual Object Tracking of Multiple Objects for Robotics" (22 citations), Wilson addressed critical failures in visual tracking—such as object occlusion, collision, and identical appearances—by leveraging audio cues to maintain robust tracking. He further advanced the field with his 2019 work on "Analyzing Liquid Pouring Sequences via Audio-Visual Neural Networks" (17 citations), where he introduced novel multimodal CNNs that estimate poured liquid weight using sound alone, eliminating the need for predefined source weights. Most recently, his 2022 study on "Audio-Visual Depth and Material Estimation for Robot Navigation" (3 citations) tackles the challenge of reflective and textureless surfaces in scene reconstruction, using sound reflections to improve depth estimation and material classification. Wilson’s contributions are reshaping how robots perceive and interact with complex, dynamic environments, making him a key figure in the evolution of robust, multimodal robotic systems.