This lecture covers the concept of computational active vision, focusing on the integration of visual intelligence with sensorimotor contingencies. Topics include mid-level visual representations, visual navigation tasks, object rearrangement, and embodied vision-and-language challenges. The instructor discusses the importance of perceptual priors in improving generalization and sample complexity for learning visuomotor policies, along with practical applications in interactive and social navigation scenarios.
This video is available exclusively on Mediaspace for a restricted audience. Please log in to MediaSpace to access it if you have the necessary permissions.
Watch on Mediaspace