Research map / 3D vision / Depth and stereo
End-to-End Learning of Keypoint Detector and Descriptor for Pose Invariant 3D Matching
Georgios Georgakis, Srikrishna Karanam, Ziyan Wu, Jan Ernst, Jana Košecká
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 the score loss to train the detector alongside contrastive descriptor learning (ablation) | Ours · Engine noisy | matching accuracy (%): 11.1 -> 23.8 | Improved |
| Add the score loss to train the detector alongside contrastive descriptor learning (ablation) | Ours · Stanford 3D models noisy, average | matching accuracy (%): 16.5 -> 32.6 | Improved |
| Select keypoints by learned scores rather than randomly (ablation) | Ours · Engine noisy | matching accuracy (%): 7.3 -> 23.8 | Improved |
| Train on noisy rather than noise-free views (ablation) | Ours · Engine noisy | matching accuracy (%): 17.8 -> 23.8 | Improved |
| Jointly learn keypoint detection and descriptors instead of using the strongest listed baseline | Ours · Stanford 3D models noisy, average | matching accuracy (%): 18.8 -> 32.6 | Improved |
| Jointly learn keypoint detection and descriptors instead of using the strongest listed baseline | Ours · MSR-7 | matching accuracy (%): 38.2 -> 41.2 | Improved |
| Use the learned method instead of Harris3D+FPFH | Ours · Engine noise-free | matching accuracy (%): 79.1 -> 67.4 | Hurt |
Closest papers by meaning
- Discovery of Latent 3D Keypoints via End-to-end Geometric Reasoning (NeurIPS 2018 · similarity 0.90)
- Learning 3D Keypoint Descriptors for Non-Rigid Shape Matching (ECCV 2018 · similarity 0.89)
- Learning and Matching Multi-View Descriptors for Registration of Point Clouds (ECCV 2018 · similarity 0.88)
- Hierarchical Metric Learning and Matching for 2D and 3D Geometric Correspondences (ECCV 2018 · similarity 0.88)
- Reinforced Feature Points: Optimizing Feature Detection and Description for a High-Level Task (CVPR 2020 · similarity 0.88)
- P2-Net: Joint Description and Detection of Local Features for Pixel and Point Matching (ICCV 2021 · 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.
