Presentation

<
51. See/Saw: Embodied Probing of Vision Transformer Attention in Virtual Reality
SessionPosters: Augmented & Virtual Reality
DescriptionSee/Saw is a VR system that paints a frozen vision transformer's patch groupings and attention onto the 3D surfaces a user aims at in a photorealistic scene, so that turning and walking re-tokenizes the view and the painted regions visibly redraw on the geometry the model saw.

INVITED TO THE FIRST ROUND OF THE STUDENT RESEARCH COMPETITION
Contributors
Ziyan Xie
University of California Los Angeles
Sihui Lin
University of California Los Angeles
Event Type
Poster
TimeSunday, 19 July 20269:00am - 5:30pm PDT
LocationWest Hall Lobby
Session TimeSunday, 19 July 20269:00am - 5:30pm PDTMonday, 20 July 20269:00am - 5:30pm PDTTuesday, 21 July 20269:00am - 5:30pm PDTWednesday, 22 July 20269:00am - 5:30pm PDTThursday, 23 July 20269:00am - 5:30pm PDT
LocationWest Hall Lobby
Registration Categories
Full Conference Supporter
Full Conference
Experience
Next PresentationNext Presentation52. See What I Hear, Hear What You See: Asymmetric Cross-Modal Play in a Cooperative VR Game
Similar Presentations
From Human Vision to Machine Vision
91. SceneTrust: Real-Time Confidence-Aware Scene Rendering for Transparent Autonomous Vehicle Decision Visualization in VR
Infant Vision: Reconfiguring Perception in Mixed Reality