3D vision has made tremendous progress in the past ten years. Autonomous cars with LIDARs collect petabytes of 3D data, and depth sensors and repositories like Shapenet provide the millions of instances required by large-scale learning systems. At the same time, a wide variety of classic 3D vision problems, ranging from 3D estimation from still images to 3D model retrieval have seen dramatic performance gains, enabling new capabilities.

Unfortunately, despite this progress, there is a fundamental disconnect between most 3D vision and the rest of machine intelligence. In computer vision, 3D papers typically focus on 3D for its own sake, and non-3D papers ignore the underlying 3D structure of the world. Outside of vision, there are even larger disconnects.

The goal of this workshop is to help bridge this gap. We want to examine how 3D vision can help and be helped by other areas of machine intelligence. Our goal is to highlight work that tries to join 3D vision and other disciplines, bring together experts in multiple domains, and to encourage discussion of how to connect with 3D vision. We are particularly interested in work that joins 3D vision with robotics, graphics, cognition, NLP, virtual reality, and semantics.


We will have more information closer to CVPR. But Bridges to 3D will be an all-day affair featuring an exciting line-up speakers and invited posters from both CVPR18/ECCV18 and non-CVPR/ICCV/ECCV conferences such as SIGGRAPH and ICRA. For an idea of how this might work, see our site from last year.



David Fouhey

Qixing Huang

Joseph Lim

University of California, Berkeley
University of Texas at Austin
University of Southern California

Hao Su

Shubham Tulsiani

University of California, San Diego
University of California, Berkeley
Senior Organizers

Jitendra Malik

David Forsyth

University of California, Berkeley
University of Illinois, Urbana-Champaign

Contact Info

E-mail: {dfouhey,shubhtuls} at eecs dot berkeley dot edu