Research map / 3D vision / Face and shape
Scalar Function Topology Divergence: Comparing Topology of 3D Objects
Ilya Trofimov, Daria Voronkova, Eduard Tulchinskii, Evgeny Burnaev, Serguei Barannikov
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 |
|---|---|---|---|
| replace Wasserstein-distance topological loss with SFTD loss (dims 0-3) in SHAPR | SHAPR · Red blood cell 3D reconstruction | Volume error: 0.291 -> 0.266 | Improved |
| replace Wasserstein-distance topological loss with SFTD loss (dims 0-2) in SHAPR | SHAPR · Red blood cell 3D reconstruction | 1-IoU: 0.467 -> 0.449 | Improved |
| include H3 (dimensions up to 3) in SFTD loss instead of up to 2 (ablation) | SHAPR+SFTD · Red blood cell 3D reconstruction | W2 distance H2: 0.065 -> 0.025 | Improved |
| add SFTD loss to SHAPR vs SHAPR with W.D. | SHAPR · Cell nuclei 3D reconstruction | 1-IoU: 0.612 -> 0.618 (dims 0-2), 0.612 -> 0.597 (dims 0-3) | Mixed |
| replace Betti matching loss with SFTD loss | 2D segmentation model · CREMI 2D segmentation | beta err: 79.16 -> 62.64 | Improved |
| replace Betti matching loss with SFTD loss | 2D segmentation model · Colon 2D segmentation | tau err: 32.00 -> 21.50 | Improved |
| replace Betti matching loss with SFTD loss | 2D segmentation model · CREMI 2D segmentation | Dice: 0.893 -> 0.885 | Hurt |
Closest papers by meaning
- Topology-Aware Segmentation Using Discrete Morse Theory (ICLR 2021 · similarity 0.87)
- Spot the Difference: Detection of Topological Changes via Geometric Alignment (NeurIPS 2021 · similarity 0.87)
- Conformable Convolution for Topologically Constrained Learning of Complex Anatomical Structures (AAAI 2026 · similarity 0.87)
- Topograph: An Efficient Graph-Based Framework for Strictly Topology Preserving Image Segmentation (ICLR 2025 · similarity 0.87)
- Diffeomorphic interpolation for efficient persistence-based topological optimization (NeurIPS 2024 · 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.
