Research map / 3D vision / Depth and stereo
T-Net: Effective Permutation-Equivariant Network for Two-View Correspondence Learning
Zhen Zhong, Guobao Xiao, Linxin Zheng, Yan Lu, Jiayi Ma
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 T-Structure feature concatenation across sub-networks (ablation) | iterative network with PCSE · YFCC100M unknown scenes | mAP@5deg (%): 47.73 -> 52.28 | Improved |
| use T-Net instead of OA-Net++ (SIFT, no RANSAC) | T-Net · YFCC100M unknown scenes | mAP@5deg (%): 38.95 -> 52.28 | Improved |
| use T-Net instead of OA-Net++ (SIFT, no RANSAC) | T-Net · SUN3D unknown scenes | mAP@5deg (%): 16.18 -> 19.71 | Improved |
| add RANSAC post-processing to T-Net (SIFT) | T-Net · SUN3D unknown scenes | mAP@5deg (%): 19.71 -> 18.00 | Hurt |
| add RANSAC post-processing to T-Net (SIFT) | T-Net · YFCC100M unknown scenes | mAP@5deg (%): 52.28 -> 56.08 | Improved |
| use SuperPoint instead of SIFT features for T-Net (no RANSAC) | T-Net · YFCC100M unknown scenes | mAP@5deg (%): 52.28 -> 40.62 | Hurt |
| use SuperPoint instead of SIFT features for OA-Net++ (no RANSAC) | OA-Net++ · SUN3D unknown scenes | mAP@5deg (%): 16.18 -> 15.62 | No clear effect |
| use T-Net instead of SuperGlue without RANSAC | T-Net · SUN3D unknown scenes | mAP@5deg (%): 6.30 -> 14.45 | Improved |
| use T-Net instead of SuperGlue with RANSAC | T-Net · SUN3D unknown scenes | mAP@5deg (%): 16.09 -> 16.35 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- OA-Net++ — OANet: Learning Two-View Correspondences and Geometry Using Order-Aware Network (TPAMI 2020) · better on 9 of 9 shared results
- SuperGlue — SuperGlue: Learning Feature Matching With Graph Neural Networks (CVPR 2020) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- T-Net++: Effective Permutation-Equivariance Network for Two-View Correspondence Pruning (TPAMI 2024 · similarity 0.94)
- Learning Two-View Correspondences and Geometry Using Order-Aware Network (ICCV 2019 · similarity 0.90)
- OANet: Learning Two-View Correspondences and Geometry Using Order-Aware Network (TPAMI 2020 · similarity 0.90)
- BCLNet: Bilateral Consensus Learning for Two-View Correspondence Pruning (AAAI 2024 · similarity 0.88)
- Multiview Equivariance Improves 3D Correspondence Understanding with Minimal Feature Finetuning (ICLR 2025 · similarity 0.85)
- MGNet: Learning Correspondences via Multiple Graphs (AAAI 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.
