Research map / 3D vision / Face and shape
XSurfer: Reconstructing surface meshes of cerebral and cerebellar cortex from diverse MRI data using untrained neural networks
Haoxiang Li, Mingxuan Liu, Divya Varadarajan, Zhangxuan Hu, Qiyuan Tian, Jonathan Polimeni
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 unidirectional instead of bi-directional Chamfer loss (ablation) | XSurfer · UKB T1 | ASSD (mm): 0.416 -> 0.303 | Improved |
| add inflation loss (ablation) | XSurfer · UKB T1 | ASSD (mm): 0.381 -> 0.303 | Improved |
| add CSF recovery segmentation refinement (ablation) | XSurfer · UKB T1 | ASSD (mm): 0.407 -> 0.303 | Improved |
| add spring loss (ablation) | XSurfer · UKB T1 | ASSD (mm): 0.332 -> 0.303 | Improved |
| add spring loss (ablation) | XSurfer · UKB T1 | HD90 (mm): 0.711 -> 0.763 | Hurt |
| add pre-training on 100 subjects (direct inference) (ablation) | XSurfer · UKB T1 | ASSD (mm): 0.303 -> 0.346 | Hurt |
| initialize TTO from pre-trained weights (ablation) | XSurfer · UKB T1 | ASSD (mm): 0.303 -> 0.347 | Hurt |
| add untrained network prior to direct optimization (ablation) | P3D-Fitting · UKB T1 | ASSD (mm): 0.667 -> 0.303 | Improved |
| add surface deformation after segmentation refinement (ablation) | Refined MC · UKB T1 | ASSD (mm): 0.411 -> 0.303 | Improved |
| add segmentation refinement to marching cubes (ablation) | Marching Cubes · UKB T1 | ASSD (mm): 0.592 -> 0.411 | Improved |
| add surface deformation after segmentation refinement (ablation) | Refined MC · Marmoset | ASSD (mm): 0.123 -> 0.091 | Improved |
| add surface deformation (ablation) | Marching Cubes · dHCP Fetal | ASSD (mm): 0.915 -> 0.717 | Improved |
Closest papers by meaning
- Coupled Reconstruction of Cortical Surfaces by Diffeomorphic Mesh Deformation (NeurIPS 2023 · similarity 0.88)
- Vox2Cortex: Fast Explicit Reconstruction of Cortical Surfaces From 3D MRI Scans With Geometric Deep Neural Networks (CVPR 2022 · similarity 0.87)
- Voxurf: Voxel-based Efficient and Accurate Neural Surface Reconstruction (ICLR 2023 · similarity 0.85)
- FlatClip: Reusing Image Foundation Models for fMRI Representation Learning via Cortical Flatmaps (NeurIPS 2026 · similarity 0.83)
- MosaicMRI: A Diverse Dataset and Benchmark for Raw Musculoskeletal MRI (NeurIPS 2026 · similarity 0.82)
- Unraveling Normal Anatomy via Fluid-Driven Anomaly Randomization (CVPR 2025 · similarity 0.82)
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.
