Research map / Graph learning / Graph algorithms and clustering
IA-GM: A Deep Bidirectional Learning Method for Graph 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 IA-GM instead of PCA-GM | Graph matching on PASCAL Keypoints · PASCAL Keypoints cross-class | accuracy: 63.90 -> 66.58 | Helped |
| Use IA-GM instead of PCA-GM | Graph matching on PASCAL Keypoints · PASCAL Keypoints in-class | accuracy: 65.24 -> 68.84 | Helped |
| Use IA-GM instead of FGM-D | Directed graph matching · Cars | accuracy: 89.7 -> 94.8 | Helped |
| Use IA-GM instead of FGM-D | Directed graph matching · Motorbikes | accuracy: 95.5 -> 92.6 | Hurt |
| Use IA-GM instead of FGM-D | Directed graph matching · Cars/Motorbikes mean | accuracy: 92.6 -> 93.7 | Helped |
| Use IA-GM instead of the strongest listed unsupervised baseline | Unsupervised graph matching · CMU House | accuracy: 99.8 -> 99.5 | Hurt |
| Use IA-GM instead of the strongest listed unsupervised baseline | Unsupervised graph matching · CMU Hotel | accuracy: 94.8 -> 97.8 | Helped |
| Use IA-GM instead of the strongest listed unsupervised baseline | Unsupervised graph matching · CMU House/Hotel mean | accuracy: 97.3 -> 98.7 | Helped |
| Replace the fully connected layer with SR-GGNN (ablation) | IA-GM ablation baseline · PASCAL Keypoints cross-class ablation | accuracy: 58.95 -> 63.04 | Helped |
| Add feature enrichment and fusion using matching feedback (ablation) | SR-GGNN graph-matching pipeline · PASCAL Keypoints cross-class ablation | accuracy: 63.04 -> 66.10 | Helped |
| Learn the similarity matrix between graphs (ablation) | SR-GGNN with feature enrichment · PASCAL Keypoints cross-class ablation | accuracy: 66.10 -> 66.32 | Helped |
| Learn the probability matrix (ablation) | SR-GGNN with similarity learning and feature enrichment · PASCAL Keypoints cross-class ablation | accuracy: 66.32 -> 66.58 | Helped |
Closest papers by meaning
- Self-Supervised Bidirectional Learning for Graph Matching (AAAI 2023 · similarity 0.94)
- Deep Learning of Graph Matching (CVPR 2018 · similarity 0.91)
- Learning deep graph matching with channel-independent embedding and Hungarian attention (ICLR 2020 · similarity 0.91)
- Learning Partial Graph Matching via Optimal Partial Transport (ICLR 2025 · similarity 0.90)
- Discrete Cycle-Consistency Based Unsupervised Deep Graph Matching (AAAI 2024 · similarity 0.87)
- M3C: A Framework towards Convergent, Flexible, and Unsupervised Learning of Mixture Graph Matching and Clustering (ICLR 2024 · similarity 0.86)
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.