AI Paper Atlas

Research map / 3D vision / Depth and stereo

RobustSpring: Benchmarking Robustness to Image Corruptions for Optical Flow, Scene Flow and Stereo

Victor Oei, Jenny Schmalfuss, Lukas Mehl, Madlen Bartsch, Shashank Agnihotri, Margret Keuper, Andreas Bulling, Andres Bruhn

ICLR 2026 · 3D vision › Depth and stereo · approach: Learned optical and scene flow

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
exclude rain pixels (about 90% of image) from the robustness score (ablation)RobustSpring metric · RobustSpring optical flowR^c_EPE rain: score changes by at most 5%No clear effect
use RobustSpring subsampling (0.05%) instead of Spring's original 1% subsampling (ablation)RobustSpring metric · RobustSpring optical flowR^c_EPE: SEA-RAFT 3.19 -> 2.96 (matches full data 2.96)Improved
subsample Hero-frames in addition (ablation)RobustSpring metric · RobustSpring optical flowR^c_EPE: Spring 3.19 vs Spring* 2.96 for SEA-RAFTImproved
use average vs median vs Schulze ranking of models (ablation)RobustSpring metric · RobustSpring optical flowmodel ranking: GMFlow rank 2 (Average) vs 5 (Median, Schulze)Mixed
more accurate optical flow modelsoptical flow models · RobustSpringR^c_EPE: trend weak and nonlinearMixed
more accurate optical flow models under noise corruptionsoptical flow models · RobustSpring noiseR^c_EPE noiseHurt
more accurate optical flow models under weather corruptionsoptical flow models · RobustSpring weatherR^c_EPE weatherImproved
use transformer architecture (GMFlow, FlowFormer)optical flow · RobustSpring noiseR^c_EPE noiseHurt
use stacked/progressive refinement (SEA-RAFT, FlowNet2)optical flow · RobustSpring noiseR^c_EPE noise: SEA-RAFT Gaussian noise 1.49; FlowNet2 1.33Improved
use scene flow models instead of conventional optical flowscene flow · RobustSpringR^c_EPENo clear effect
evaluate on noisier real KITTI frames instead of RobustSpring noiseoptical flow · KITTI noisiest 10%relative noise robustness: rankings transfer except FlowFormerImproved
use M-FUSE rather than RAFT-3D for optical flowscene flow · RobustSpringAverage R^c_EPE: 5.03 -> 3.39Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy