RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching

09/15/2021

∙

by   Lahav Lipson, et al.

∙

0

∙

share

We introduce RAFT-Stereo, a new deep architecture for rectified stereo based on the optical flow network RAFT. We introduce multi-level convolutional GRUs, which more efficiently propagate information across the image. A modified version of RAFT-Stereo can perform accurate real-time inference. RAFT-stereo ranks first on the Middlebury leaderboard, outperforming the next best method on 1px error by 29 stereo benchmark. Code is available at https://github.com/princeton-vl/RAFT-Stereo.