Research map / Detection and segmentation / Open-set and few-shot detection
An Alarm System for Segmentation Algorithm Based on Shape Model
Fengze Liu, Yingda Xia, Dong Yang, Alan L. Yuille, Daguang Xu
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 VAE shape feature (128-d) instead of direct regression | quality alarm · NIH | MAE: 6.30 -> 2.89 | Improved |
| use VAE shape feature (128-d) instead of uncertainty (Kwon et al.) | quality alarm · MSD | MAE: 12.68 -> 8.14 | Improved |
| use VAE shape feature (128-d) instead of uncertainty (Jungo et al.) | quality alarm · Synapse | MAE: 9.45 -> 4.93 | Improved |
| increase VAE latent dimension from 2 to 128 (ablation) | VAE shape feature · NIH | MAE: 5.31 -> 2.89 | Improved |
| increase VAE latent dimension from 16 to 128 (ablation) | VAE shape feature · NIH | MAE: 4.39 -> 2.89 | Improved |
| increase VAE latent dimension from 128 to 1024 (ablation) | VAE shape feature · NIH | MAE: 2.89 -> 3.50 | Hurt |
| increase VAE latent dimension from 128 to 1024 (ablation) | VAE shape feature · Synapse | MAE: 4.93 -> 5.71 | Hurt |
| increase VAE latent dimension from 128 to 1024 (ablation) | VAE shape feature · MSD | MAE: 8.14 -> 8.42 | Hurt |
| concatenate image with mask in direct regression (ablation) | Direct Regression · NIH | MAE: 6.30 -> 11.74 | Hurt |
| concatenate image with mask in direct regression (ablation) | Direct Regression · MSD | MAE: 14.47 -> 21.87 | Hurt |
| evaluate tumor (weak shape prior) rather than pancreas | VAE-1024 alarm · MSD | MAE: 6.03 -> 20.21 | Hurt |
| use VAE-1024 alarm instead of Kwon et al. on tumor | quality alarm · MSD tumor | MAE: 26.14 -> 20.21 | Improved |
Closest papers by meaning
- Amodal Segmentation Based on Visible Region Segmentation and Shape Prior (AAAI 2021 · similarity 0.82)
- ShapeMask: Learning to Segment Novel Objects by Refining Shape Priors (ICCV 2019 · similarity 0.81)
- On Image Segmentation With Noisy Labels: Characterization and Volume Properties of the Optimal Solutions to Accuracy and Dice (NeurIPS 2022 · similarity 0.81)
- COMPASS: Robust Feature Conformal Prediction for Medical Segmentation Metrics (ICLR 2026 · similarity 0.81)
- MCMC Shape Sampling for Image Segmentation With Nonparametric Shape Priors (CVPR 2016 · similarity 0.81)
- Interactive Medical Image Segmentation: A Benchmark Dataset and Baseline (CVPR 2025 · similarity 0.80)
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.
