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
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add Multi-Level Context Ultra-Aggregation to PSM-Net | PSM-Net · KITTI 2015 test | D1-all/est: 2.32 -> 2.09 | Improved |
| replace many 3D convolutions with SGA and LGA layers | 4D cost volume regularization (GANet) · KITTI 2015 test | D1-all/est: 2.03 | Improved |
| regularize cost maps with a GRU along depth instead of 3D convolutions | multi-view stereo cost regularization · memory | GPU memory: 15.4GB -> around 5GB | Improved |
| use a learned decision network instead of correlation for similarity | patch-based stereo matching · stereo matching | accuracy and speed | Mixed |
| add residual blocks with multilevel weighted skip connections | Zbontar et al. matching network · stereo matching | matching accuracy | Improved |
| train with weak supervision (MIL, contrastive, contrastive-DP) instead of full labels | MC-CNN-fst architecture · KITTI/Middlebury | matching accuracy | No clear effect |
| use hybrid group-wise correlation plus concatenation volume | 3D-4D cost volume (Guo et al.) · KITTI 2015 test | D1-all/est: 2.11 | Improved |
| learn SGM penalties with SGM-Net instead of hand tuning | semi-global matching · stereo matching | disparity accuracy | Improved |
| compress 4D cost volume into compact matching signatures | Tulyakov et al. · stereo matching | memory | Improved |
| use coarse-to-fine residual cost volumes | AnyNet · stereo matching | accuracy vs speed | Improved |
Closest papers by meaning
- On the Synergies between Machine Learning and Binocular Stereo for Depth Estimation from Images: a Survey (TPAMI 2021 · similarity 0.94)
- On the confidence of stereo matching in a deep-learning era: a quantitative evaluation (TPAMI 2021 · similarity 0.93)
- Learning Stereo from Single Images (ECCV 2020 · similarity 0.91)
- Guided Stereo Matching (CVPR 2019 · similarity 0.91)
- Stereo Matching by Training a Convolutional Neural Network to Compare Image Patches (JMLR 2016 · similarity 0.91)
- Stereo Risk: A Continuous Modeling Approach to Stereo Matching (ICML 2024 · similarity 0.90)
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.
