Research map / Detection and segmentation / CNN features
Combinatorial Energy Learning for Image Segmentation
Jeremy B Maitin-Shepard, Viren Jain, Michal Januszewski, Peter Li, Pieter Abbeel
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 CELIS combinatorial energy agglomeration instead of GALA | CELIS · FIB-25 11-gigavoxel test set | Variation of Information and Rand F1 | Improved |
| use CELIS instead of GALA on 18 subvolumes | CELIS · 18 FIB-25 500^3 subvolumes | Variation of Information: 19% median reduction | Improved |
| use CELIS instead of GALA on 18 subvolumes | CELIS · 18 FIB-25 500^3 subvolumes | Rand F1 error: 22% median reduction | Improved |
| CELIS vs GALA on subvolume (2428, 2856, 6890) | CELIS · FIB-25 subvolume (2428, 2856, 6890) | Variation of Information: 1.380077 -> 0.869191 | Improved |
| CELIS vs GALA on subvolume (3056, 2856, 5006) | CELIS · FIB-25 subvolume (3056, 2856, 5006) | Variation of Information: 0.902916 -> 0.676395 | Improved |
| CELIS vs GALA on subvolume (2428, 2856, 5006) | CELIS · FIB-25 subvolume (2428, 2856, 5006) | Variation of Information: 0.550208 -> 0.543177 | Improved |
| CELIS vs GALA on subvolume (3056, 2228, 6890) | CELIS · FIB-25 subvolume (3056, 2228, 6890) | Rand F1: 0.844851 -> 0.854900 | Mixed |
| CELIS vs GALA across all 18 subvolumes | CELIS · FIB-25 subvolumes | both metrics | Improved |
| CELIS overall volumetric reconstruction accuracy vs best existing method | CELIS · FIB-25 test set | reconstruction accuracy: 20% improvement | Improved |
| use CELIS vs oracle agglomeration upper bound | CELIS · FIB-25 test set | Variation of Information | Mixed |
Closest papers by meaning
- DeepMulticut: Deep Learning of Multicut Problem for Neuron Segmentation From Electron Microscopy Volume (TPAMI 2024 · similarity 0.85)
- Fast Neural Architecture Search of Compact Semantic Segmentation Models via Auxiliary Cells (CVPR 2019 · similarity 0.85)
- Progressive Test Time Energy Adaptation for Medical Image Segmentation (ICCV 2025 · similarity 0.85)
- LoMix: Learnable Weighted Multi-Scale Logits Mixing for Medical Image Segmentation (NeurIPS 2025 · similarity 0.84)
- Learning normalized image densities via dual score matching (NeurIPS 2025 · similarity 0.84)
- E2ENet: Dynamic Sparse Feature Fusion for Accurate and Efficient 3D Medical Image Segmentation (NeurIPS 2024 · 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.
