AI Paper Atlas

Research map / 3D vision / Depth and stereo

Displacement-Invariant Cost Computation for Stereo Matching

Yiran Zhong, Charles Loop, Wonmin Byeon, Stan T. Birchfield, Yuchao Dai, Kaihao Zhang, A. Kamenev, Thomas Michael Breuel, Hongdong Li, Jan Kautz

IJCV 2022 · 3D vision › Depth and stereo · approach: Learned stereo with specialized sensors · 14 citations (OpenAlex)

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
replace 3D-convolution matching on a 4D volume (Baseline3D) with shared 2D matching net per disparityBaseline3D · SceneFlow testruntime / EPE: 0.15 -> 0.007 s; EPE 0.78 -> 1.20Mixed
add RefineNet to the 2D matching network (ablation)2D matching net · SceneFlow testEPE: 1.20 -> 1.14Improved
add RefineNet to the 2D matching network (ablation)2D matching net · SceneFlow testbad-1.0: 10.31 -> 10.16Improved
feed warped right image to RefineNet (ablation)2D + RefineNet · SceneFlow testEPE: 1.14 -> 1.08Improved
feed entropy map to RefineNet (ablation)2D + RefineNet · SceneFlow testEPE: 1.14 -> 1.09Improved
feed entropy map to RefineNet (ablation)2D + RefineNet · SceneFlow testbad-1.0: 10.16 -> 9.67Improved
use learned matching net instead of hand-crafted dot-product cost (ablation)cost computation · SceneFlow testEPE: 6.05 -> 1.20Improved
use 2D displacement-invariant cost instead of 3D convolutionBaseline3D · Middlebury 2014 train (cross-dataset)bad-1.0: 53.0 -> 51.4Improved
use 2D displacement-invariant cost instead of 3D convolutionBaseline3D · KITTI 2015 train (cross-dataset)EPE: 1.92 -> 1.63Improved
use 2D displacement-invariant cost instead of 3D convolutionBaseline3D · KITTI 2015 train (cross-dataset)bad-1.0: 46.76% -> 37.55%Improved
use 2D displacement-invariant cost instead of 3D convolutionBaseline3D · Middlebury 2014 train (cross-dataset)avgerr: 13.5 -> 6.67Improved
displacement-invariant matching instead of context-regressing 2D net (MADNet)MADNet · Random Dot StereoEPE: 51.21 -> 1.02Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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