Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
$\texttt{PRISM}$:A 3D Probabilistic Neural Representation for Interpretable Shape Modeling
Yining Jiao, Sreekalyani Bhamidi, Carlton Jude ZDANSKI, Julia S Kimbell, Andrew Prince, Cameron P Worden, Samuel Kirse, Christopher Rutter, Benjamin H Shields, William Alexander Dunn, Jisan Mahmud, Marc Niethammer
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 local (pointwise) instead of global intrinsic time for OOD scoring (ablation) | PRISM · OOD detection, Pediatric Airway | AUC: 0.502 -> 0.875 | Improved |
| use local instead of global intrinsic time scoring (ablation) | PRISM · OOD detection, Pediatric Airway | accuracy: 0.470 -> 0.857 | Improved |
| use PRISM instead of NAISR for personalized prediction | PRISM · Starman (L) personalized prediction | Chamfer distance (x100): 0.595 -> 0.099 | Improved |
| use PRISM instead of A-SDF for personalized prediction | PRISM · ANNY personalized prediction | Chamfer distance (x100): 0.081 -> 0.018 | Improved |
| use PRISM instead of A-SDF for mean reconstruction on small data | PRISM · Airway mean reconstruction | Chamfer distance (x100): 0.114 -> 0.064 | Improved |
| use PRISM instead of A-SDF for mean reconstruction on large synthetic data | PRISM · Starman (G) mean reconstruction | Chamfer distance (x100): 0.137 vs 0.138 | No clear effect |
| use PRISM instead of A-SDF for mean reconstruction | PRISM · ANNY mean reconstruction | Chamfer distance (x100): 0.040 vs 0.048 | Hurt |
| amortized inverse encoder instead of test-time optimization (ablation) | PRISM · Airway global intrinsic time | Pearson r: 0.923 -> 0.893 | Hurt |
| amortized inverse encoder instead of test-time optimization (ablation) | PRISM · Airway global intrinsic time | time per case (s): 8.838 -> 0.805 | Improved |
| use amortized PRISM instead of TTO baselines | PRISM · Starman (G) intrinsic time | MAE: 0.019 -> 0.005 | Improved |
| use mean-only Fisher information instead of full Fisher information (ablation) | PRISM · Starman (G) temporal uncertainty | uncertainty band vs ground truth | Improved |
Closest papers by meaning
- $\texttt{NAISR}$: A 3D Neural Additive Model for Interpretable Shape Representation (ICLR 2024 · similarity 0.85)
- Uncertainty Estimation for Pretrained Medical Image Registration Models via Transformation Equivariance (NeurIPS 2026 · similarity 0.84)
- Hierarchical Uncertainty Estimation for Learning-based Registration in Neuroimaging (ICLR 2025 · similarity 0.83)
- CUPS: Improving Human Pose-Shape Estimators with Conformalized Deep Uncertainty (ICML 2025 · similarity 0.83)
- Point2SSM: Learning Morphological Variations of Anatomies from Point Clouds (ICLR 2024 · similarity 0.83)
- Dynamic Neural Surfaces for Elastic 4D Shape Representation and Analysis (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.
