Research map / Label-efficient and robust learning / OOD and anomaly detection
Spot the Difference: Detection of Topological Changes via Geometric Alignment
Steffen Czolbe, Aasa Feragen, Oswin Krause
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 |
|---|---|---|---|
| Use semantic loss instead of MSE (ablation) | Conditional VAE registration · Cell topological change detection | AUC: 0.72 -> 0.88 | Improved |
| Use semantic loss instead of MSE (ablation) | Conditional VAE registration · Brain tumor detection | AUC: 0.61 -> 0.80 | Improved |
| Use the proposed semantic-loss method rather than Li and Wyatt | Unsupervised topological change detection · Cell topological change detection | AUC: 0.75 -> 0.88 | Improved |
| Use the proposed semantic-loss method rather than Li and Wyatt | Unsupervised brain tumor detection · Brain tumor detection | AUC: 0.70 -> 0.80 | Improved |
| Use the proposed semantic-loss method rather than supervised segmentation | Topological change detection · Cell topological change detection | AUC: 0.90 -> 0.88 | Hurt |
| Use the proposed semantic-loss method rather than supervised segmentation | Brain tumor detection · Brain tumor detection | AUC: 0.95 -> 0.80 | Hurt |
| Replace semantic loss with MSE (ablation) | Conditional VAE registration · Cells | Dice: 0.90 -> 0.89 | Hurt |
| Replace semantic loss with MSE (ablation) | Conditional VAE registration · Brains | Dice: 0.56 -> 0.52 | Hurt |
| Replace semantic loss with MSE (ablation) | Conditional VAE registration · Cells | Seg. Accuracy (%): 95.4 -> 95.0 | Hurt |
| Replace semantic loss with MSE (ablation) | Conditional VAE registration · Brains | Seg. Accuracy (%): 84.7 -> 84.1 | Hurt |
| Replace semantic loss with MSE (ablation) | Conditional VAE registration · Brain tumor detection | AUC: 0.80 -> 0.61 | Hurt |
Closest papers by meaning
- Topology-Aware Non-Rigid Point Cloud Registration (TPAMI 2019 · similarity 0.87)
- Scalar Function Topology Divergence: Comparing Topology of 3D Objects (ECCV 2024 · similarity 0.87)
- TopoCL: Topological Contrastive Learning for Medical Imaging (CVPR 2026 · similarity 0.87)
- Topograph: An Efficient Graph-Based Framework for Strictly Topology Preserving Image Segmentation (ICLR 2025 · similarity 0.86)
- TopoDiffusionNet: A Topology-aware Diffusion Model (ICLR 2025 · similarity 0.85)
- Towards Scalable Topological Regularizers (ICLR 2025 · similarity 0.84)
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.
