AI Paper Atlas

Research map / 3D vision / Depth and stereo

On the Synergies between Machine Learning and Binocular Stereo for Depth Estimation from Images: a Survey

Matteo Poggi, Fabio Tosi, Konstantinos Batsos, Philippos Mordohai, Stefano Mattoccia

TPAMI 2021 · 3D vision › Depth and stereo · approach: Self-supervised monocular depth · 169 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 learned matching cost (MC-CNN) over classical SGBMstereo pipeline · KITTI 2015 leaderboardD1-all error (%): 10.86 -> 3.89Improved
build learned SGM-Net optimization on MC-CNNMC-CNN-acrt · KITTI 2015 leaderboardD1-all error (%): 3.89 -> 3.66Improved
add learned disparity refinement on top of other algorithmsstereo pipeline · KITTI 2015 leaderboardD1-all error (%): 3.89 -> 3.03Improved
use end-to-end CSPN network instead of MC-CNN pipelineMC-CNN-acrt · KITTI 2015 leaderboardD1-all error (%): 3.89 -> 1.74Improved
use real-time end-to-end networks (DispNet-C, MADNet, StereoNet)end-to-end stereo · KITTI 2015 leaderboardD1-all error (%): 4.34, 4.66, 4.83 vs 3.89Hurt
use 3D cost-volume architectures instead of 2Dend-to-end stereo · KITTI 2015 leaderboardD1-all error (%): 1.74 vs 2.02Mixed
use end-to-end HSM instead of MC-CNN on high-res imagesMC-CNN-acrt · Middlebury 2014 leaderboardbad 2.0 error (%): 8.08 -> 10.2Hurt
use end-to-end HSM instead of MC-CNN on high-res imagesMC-CNN-acrt · Middlebury 2014 leaderboardaverage pixel error: 3.82 -> 2.07Improved
use end-to-end DispNet-CSS instead of SGMSGM · ETH3D leaderboardbad 1.0 error (%): 10.08 -> 2.69Improved
use end-to-end DispNet-CSS instead of SGMSGM · Middlebury 2014 leaderboardbad 2.0 error (%): 18.4 -> 22.8Hurt
use larger receptive field confidence network LGC-Net instead of O1confidence estimation · KITTI 2015 confidence (SGM disparities)AUC x10^2: 4.61 -> 3.92Improved
add cost volume cues to image and disparity cues (LAF-Net)LGC-Net · KITTI 2015 confidence (SGM disparities)AUC x10^2: 3.92 -> 3.85Improved

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