AI Paper Atlas

Research map / 3D vision / Depth and stereo

Weakly Supervised Learning of Deep Metrics for Stereo Reconstruction

Stepan Tulyakov, Anton Ivanov, Francois Fleuret

ICCV 2017 · 3D vision › Depth and stereo · approach: Dense depth prediction and completion · 46 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 uniqueness constraint (CONTRASTIVE) over MIL (ablation)semi-supervised stereo metric · KITTI'12WTA error (%): 18.45 -> 17.63Improved
combine MIL and CONTRASTIVE (ablation)semi-supervised stereo metric · KITTI'12WTA error (%): 17.63 -> 16.12Improved
add dynamic programming with continuity and ordering constraints (CONTRASTIVE-DP) (ablation)CONTRASTIVE · KITTI'12WTA error (%): 17.63 -> 14.61Improved
use all constraints (CONTRASTIVE-DP) instead of only epipolar and disparity (MIL) (ablation)MIL · KITTI'12WTA error (%): 18.45 -> 14.61Improved
DP training takes longer than CONTRASTIVE (ablation)CONTRASTIVE-DP · KITTI'12training time (hr): 30 -> 68Hurt
replace supervised training with semi-supervised MC-CNN-SSMC-CNN fst · KITTI'12WTA error (%): 15.44 -> 13.90Improved
replace supervised training with semi-supervised MC-CNN-SSMC-CNN fst · KITTI'15WTA error (%): 15.38 -> 14.08Improved
replace supervised training with semi-supervised MC-CNN-SSMC-CNN fst · MiddleburyWTA error (%): 29.94 -> 30.06No clear effect
replace supervised training with semi-supervised MC-CNN-SS in full pipelineMC-CNN fst · KITTI'12 benchmarkpipeline error (%): 2.82 -> 3.02Hurt
replace supervised training with semi-supervised MC-CNN-SS in full pipelineMC-CNN fst · KITTI'15 benchmarkpipeline error (%): 4.62 -> 4.97Hurt
replace supervised training with semi-supervised MC-CNN-SS in full pipelineMC-CNN fst · Middlebury benchmarkpipeline error (%): 9.69 -> 12.3Hurt
train metric on the same dataset as the test set instead of another dataset (ablation)MC-CNN-SS · KITTI'12 WTAWTA error (%): 15.52 -> 13.90Improved

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