AI Paper Atlas

Research map / 3D vision / Depth and stereo

Beyond local reasoning for stereo confidence estimation with deep learning

Fabio Tosi, Matteo Poggi, Antonio Benincasa, Stefano Mattoccia

ECCV 2018 · 3D vision › Depth and stereo · approach: Dense depth prediction and completion · 66 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 global ConfNet branch and cascaded fusion to CCNNCCNN with AD-CENSUS · KITTI 2012AUC: 0.1207 -> 0.1174Improved
add global ConfNet branch and cascaded fusion to CCNNCCNN with MC-CNN · KITTI 2012AUC: 0.0291 -> 0.0279Improved
add global ConfNet branch and cascaded fusion to CCNNCCNN with SGM · KITTI 2012AUC: 0.0194 -> 0.0176Improved
add global ConfNet branch and cascaded fusion to CCNNCCNN with SGM · KITTI 2015AUC: 0.0201 -> 0.0193Improved
add global ConfNet branch and cascaded fusion to CCNNCCNN with SGM · Middlebury v3 (cross-validation)AUC: 0.0794 -> 0.0703Improved
add global ConfNet branch and cascaded fusion to CCNNCCNN with AD-CENSUS · Middlebury v3 (cross-validation)AUC: 0.1131 -> 0.1099Improved
use global ConfNet alone instead of local CCNNConfNet with AD-CENSUS · KITTI 2012AUC: 0.1207 -> 0.1295Hurt
use global ConfNet alone instead of local CCNNConfNet with SGM · KITTI 2012AUC: 0.0194 -> 0.0199Hurt
replace local CCNN input with RGB-augmented early fusion EFNCCNN with SGM · KITTI 2012AUC: 0.0194 -> 0.0229Hurt
add RGB image via late fusion LFN in place of CCNNCCNN with MC-CNN · KITTI 2012AUC: 0.0291 -> 0.0294No clear effect
use LFN instead of CCNN as local network inside LGC-NetLGC-Net with MC-CNN · KITTI 2012AUC: 0.0279 -> 0.0278No clear effect
use LFN instead of CCNN as local network inside LGC-NetLGC-Net with SGM · Middlebury v3 (cross-validation)AUC: 0.0703 -> 0.0709No clear effect

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