Papers

OVTrack: Open-Vocabulary Multiple Object Tracking
We introduce the first open-vocabulary multiple object tracker OVTrack trained from only static images and an evaluation benchmark.
West Exhibit Hall ABC, Poster ID 137, Tuesday, June 20 (Afternoon) 1630–1830 PDT
[Webste] [Code] [Paper]

Mask-Free Video Instance Segmentation
We remove video and image mask annptation necessity for training highly accurate VIS models.
West Exhibit Hall ABC, Poster ID 215, Thursday, June 22 (Afternoon) 1630–1830 PDT
[Webste] [Code] [Paper]

iDisc: Internal Discretization for Monocular Depth Estimation
We propose a monocular depth estimation method which represents internally the scene as a finite set of concepts via a continuous-discrete-continuous bottleneck.
West Exhibit Hall ABC, Poster ID 083, Thursday, June 22 (Afternoon) 1630–1830 PDT
[Webste] [Code] [Paper]

Workshop Talks

2nd Workshop on Tracking and Its Many Guises
East Hall 11, Vancouver Convention Center, June 18
Fisher Yu, 10:45am - 11:15am
Martin Danelljan, 11:26am -11:37am
[Website]

Vision-Centric Autonomous Driving
West 202-204, Vancouver Convention Center, June 19
Fisher Yu, 10:15am - 11am
[Website]

Organized Workshop

Workshop on Autonomous Driving
East Ballroom C, Sunday, June 18
[Website]