AI Paper Atlas

Research map / 3D vision / Depth and stereo

A Survey on Deep Learning Techniques for Stereo-Based Depth Estimation

Hamid Laga, Laurent Valentin Jospin, Farid Boussaïd, Mohammed Bennamoun

TPAMI 2020 · 3D vision › Depth and stereo · approach: Dense depth prediction and completion · 368 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 Multi-Level Context Ultra-Aggregation to PSM-NetPSM-Net · KITTI 2015 testD1-all/est: 2.32 -> 2.09Improved
replace many 3D convolutions with SGA and LGA layers4D cost volume regularization (GANet) · KITTI 2015 testD1-all/est: 2.03Improved
regularize cost maps with a GRU along depth instead of 3D convolutionsmulti-view stereo cost regularization · memoryGPU memory: 15.4GB -> around 5GBImproved
use a learned decision network instead of correlation for similaritypatch-based stereo matching · stereo matchingaccuracy and speedMixed
add residual blocks with multilevel weighted skip connectionsZbontar et al. matching network · stereo matchingmatching accuracyImproved
train with weak supervision (MIL, contrastive, contrastive-DP) instead of full labelsMC-CNN-fst architecture · KITTI/Middleburymatching accuracyNo clear effect
use hybrid group-wise correlation plus concatenation volume3D-4D cost volume (Guo et al.) · KITTI 2015 testD1-all/est: 2.11Improved
learn SGM penalties with SGM-Net instead of hand tuningsemi-global matching · stereo matchingdisparity accuracyImproved
compress 4D cost volume into compact matching signaturesTulyakov et al. · stereo matchingmemoryImproved
use coarse-to-fine residual cost volumesAnyNet · stereo matchingaccuracy vs speedImproved

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