Meeting ID: 862 3823 3298
Title: Learning 3D from Moving Pictures
Abstract: The illusion of motion created by a rapid succession of still images seems like magic. Such motion in video -- both motion of the camera, and of objects and people -- is also very useful as a cue for machines to learn to perceive 3D scenes. This talk will explore the use of motion in large online video collections to learn about scene properties like depth, materials, object segmentations, and potential object movements, and learn to predict these properties from single images. I'll also talk about some recent work that seeks to also incorporate supervision from language into 3D reasoning.
Bio: Noah Snavely is an associate professor of Computer Science at Cornell University and Cornell Tech, and also a research scientist at Google. Noah's research interests are in computer vision and graphics, in particular 3D understanding and depiction of scenes from images. Noah is the recipient of a PECASE, an Alfred P. Sloan Fellowship, a SIGGRAPH Significant New Researcher Award, and a Helmholtz Prize.