RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching

15 Sep 2021  ·  Lahav Lipson, Zachary Teed, Jia Deng ·

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% and outperforms all published work on the ETH3D two-view stereo benchmark. Code is available at https://github.com/princeton-vl/RAFT-Stereo.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Stereo Disparity Estimation Middlebury 2014 RAFT-Stereo D1 Error (2px) 0.0474 # 1
Stereo Depth Estimation Spring RAFT-Stereo 1px total 15.273 # 2


No methods listed for this paper. Add relevant methods here