AI Paper Atlas

Research map / 3D vision / Depth and stereo

DeepPruner: Learning Efficient Stereo Matching via Differentiable PatchMatch

Shivam Duggal, Shenlong Wang, Wei-Chiu Ma, Rui Hu, Raquel Urtasun

ICCV 2019 · 3D vision › Depth and stereo · approach: Cost-volume multi-view stereo · 339 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
add confidence range predictor (ablation)DeepPruner-Best · KITTI 2015 validation (All)D1 outlier rate (%): 2.28 -> 1.90Improved
add confidence range predictor (ablation)DeepPruner-Best · SceneFlow testEPE: 0.982 -> 0.868Improved
add RefineNet (ablation)DeepPruner-Best · KITTI 2015 validation (All)D1 outlier rate (%): 2.32 -> 1.8Improved
add RefineNet (ablation)DeepPruner-Best · SceneFlow testEPE: 1.283 -> 0.858Improved
add second PatchMatch (PM-2) during SceneFlow pre-training (ablation)DeepPruner-Best · SceneFlow testEPE: 0.868 -> 0.858Improved
add second PatchMatch (PM-2) during SceneFlow pre-training (ablation)DeepPruner-Best · KITTI 2015 validation (All)D1 outlier rate (%): 1.90 -> 1.8Improved
add confidence range predictor (runtime cost) (ablation)DeepPruner-Best · KITTI inferenceruntime: 120 ms -> 172 msHurt
downsample cost volume by 8 instead of 4 (Fast vs Best)DeepPruner · SceneFlow testEPE and runtime: 0.86 -> 0.97 EPE, 182 ms -> 62 msMixed
prune disparity search space with PatchMatch and confidence range instead of full cost volumePSM-Net style full cost volume · KITTI full-size inferencememory (MB): 4351 MB -> 1161 MBImproved
remove most uncertain pixels using predicted rangeDeepPruner · KITTI validationoutlier ratio: removing 6% uncertain pixels improves outlier ratio by 38%Improved
DeepPruner-Fast vs prior real-time DispNetCDeepPruner-Fast · KITTI 2015 test (All)D1 outlier rate (%): 4.34 -> 2.59Improved

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