RAFT-Stereo: Multilevel Recurrent Field Transforms for Stereo Matching
arXiv:2109.07547
Abstract
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.
References in corpus (7)
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Decoupled Weight Decay Regularization
- Instance Normalization: The Missing Ingredient for Fast Stylization
- Computing the Stereo Matching Cost with a Convolutional Neural Network
- Continuous 3D Label Stereo Matching using Local Expansion Moves
- AMNet: Deep Atrous Multiscale Stereo Disparity Estimation Networks
- TartanAir: A Dataset to Push the Limits of Visual SLAM