AI Paper Atlas

Research map / 3D vision / Depth and stereo

On the confidence of stereo matching in a deep-learning era: a quantitative evaluation

Matteo Poggi, Seungryong Kim, Fabio Tosi, Sunok Kim, Filippo Aleotti, Dongbo Min, Kwanghoon Sohn, Stefano Mattoccia

TPAMI 2021 · 3D vision › Depth and stereo · approach: Learned stereo with specialized sensors · 56 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
train learned confidence measure on real KITTI 2012 instead of synthetic DrivingLAF · KITTI 2012AUC (x100, lower better): 6.87 -> 5.33Improved
train on KITTI 2012 instead of DrivingMPN · KITTI 2012AUC (x100, lower better): 6.53 -> 5.58Improved
train on KITTI 2012 instead of DrivingMPN · ETH3DAUC (x100, lower better): 7.08 -> 6.23Improved
train on KITTI 2012 instead of DrivingLAF · ETH3DAUC (x100, lower better): 9.12 -> 9.50Hurt
train on KITTI 2012 instead of DrivingPBCP_d · MiddleburyAUC (x100, lower better): 14.24 -> 15.01Hurt
use learned measure LAF instead of best hand-crafted DA31DA31 · KITTI 2015AUC (x100, lower better): 6.45 -> 5.35Improved
use learned measure FA instead of hand-crafted DA31DA31 · MiddleburyAUC (x100, lower better): 12.92 -> 13.99Hurt
use learned measure LAF instead of DA31DA31 · DrivingAUC (x100, lower better): 23.12 -> 13.76Improved
use WTA selection instead of soft-argmax in GANetGANet · KITTI 2012 (bad3)bad error (%): 9.51% -> 8.47%Improved
use WTA selection instead of soft-argmax in GANetGANet · ETH3D (bad1)bad error (%): 8.73% -> 10.80%Hurt

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