Research map / Architectures and efficiency / Neural architecture search
Hierarchical Neural Architecture Search for Deep Stereo Matching
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| use the smaller candidate operation set instead of the larger set (ablation) | LEAStereo architecture search · SceneFlow test | EPE: 1.55px -> 0.86px | Helped |
| jointly search the Feature Net and Matching Net instead of searching separately (ablation) | LEAStereo architecture search · SceneFlow test | EPE: 0.86px -> 0.78px; 9.30% | Helped |
| use residual cells instead of direct cells (ablation) | LEAStereo · SceneFlow test | EPE: 0.91px -> 0.78px; 14.29% | Helped |
| remove dilated separable convolutions and pooling layers from the feature-net candidate set (ablation) | LEAStereo architecture search · SceneFlow | performance | No clear effect |
| remove commonly used operations from the matching-net candidate set (ablation) | LEAStereo architecture search · SceneFlow | performance | No clear effect |
| use the full task-specific stereo pipeline and hierarchical full-network search rather than cell-only direct disparity regression | AutoDispNet-CSS · KITTI 2012 and KITTI 2015 | benchmark error: 1.70% -> 1.13%; 2.18% -> 1.65% | Helped |
Closest papers by meaning
- CFNet: Cascade and Fused Cost Volume for Robust Stereo Matching (CVPR 2021 · similarity 0.90)
- AANet: Adaptive Aggregation Network for Efficient Stereo Matching (CVPR 2020 · similarity 0.90)
- Efficient Deep Learning for Stereo Matching (CVPR 2016 · similarity 0.90)
- Local Similarity Pattern and Cost Self-Reassembling for Deep Stereo Matching Networks (AAAI 2022 · similarity 0.90)
- Hierarchical Object-Aware Dual-Level Contrastive Learning for Domain Generalized Stereo Matching (NeurIPS 2024 · similarity 0.89)
- Stereo Risk: A Continuous Modeling Approach to Stereo Matching (ICML 2024 · similarity 0.88)
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.