Research map / 3D vision / Depth and stereo
GOCor: Bringing Globally Optimized Correspondence Volumes into Your Neural Network
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 |
|---|---|---|---|
| replace feature correlation with GOCor | GLU-Net · HPatches | AEPE: 25.05 -> 20.16 | Helped |
| replace feature correlation with GOCor | GLU-Net · MegaDepth | PCK-1 (%): 21.58 -> 37.28 | Helped |
| replace feature correlation with GOCor | GLU-Net · KITTI-2015 | F1 (%): 33.83 -> 27.57 | Helped |
| replace local correlation with GOCor | PWC-Net (ft Sintel) · KITTI-2015 | F1 (%): 24.35 -> 20.93 | Helped |
| replace local correlation with GOCor | PWC-Net (ft 3D-Things) · KITTI-2012 | AEPE: 4.34 -> 4.12 | Helped |
| replace local correlation with GOCor | PWC-Net (ft Sintel) · Sintel Clean (training data) | AEPE: 1.70 -> 1.74 | No clear effect |
| replace feature correlation with GOCor | GLU-Net · TSS semantic matching | PCK All (%): 79.2 -> 83.4 | Helped |
| replace feature correlation with GOCor | GLU-Net · KITTI-2012 | run time (ms): 154.97 -> 261.90 | Hurt |
| add NC-Net post-processing to correlation volume (ablation) | BaseNet · HPatches | AEPE: 30.94 -> 39.15 | Hurt |
| use robust reference objective Lr instead of linear regression objective (ablation) | BaseNet + Global-GOCor · HPatches | AEPE: 27.02 -> 26.27 | Helped |
| add reference objective Lr (global GOCor) (ablation) | BaseNet · KITTI-2015 | F1 (%): 48.66 -> 46.24 | Helped |
| add query objective Lq (ablation) | BaseNet + Global-GOCor Lr · KITTI-2015 | F1 (%): 46.24 -> 43.08 | Helped |
Closest papers by meaning
- FC-GNN: Recovering Reliable and Accurate Correspondences from Interferences (CVPR 2024 · similarity 0.88)
- Unifying Feature and Cost Aggregation with Transformers for Semantic and Visual Correspondence (ICLR 2024 · similarity 0.86)
- ETO:Efficient Transformer-based Local Feature Matching by Organizing Multiple Homography Hypotheses (NeurIPS 2024 · similarity 0.86)
- RoMa: Robust Dense Feature Matching (CVPR 2024 · similarity 0.85)
- MATCHA: Towards Matching Anything (CVPR 2025 · similarity 0.85)
- GIM: Learning Generalizable Image Matcher From Internet Videos (ICLR 2024 · similarity 0.85)
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.