Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Uncertainty Estimation for Safety-critical Scene Segmentation via Fine-grained Reward Maximization
Hongzheng Yang, Cheng Chen, Yueyao Chen, Markus Scheppach, Hon Chi Yip, Qi Dou
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 FGRM instead of NatPN | FGRM · CholecSeg8K LC | ECE: 11.74 -> 9.63 | Improved |
| Use FGRM instead of NatPN | FGRM · ESD dataset | ECE: 13.67 -> 10.42 | Improved |
| Use FGRM instead of NatPN | FGRM · CholecSeg8K LC | PR: 1.56 -> 1.85 | Improved |
| Use FGRM instead of NatPN | FGRM · ESD dataset | PR: 1.45 -> 1.78 | Improved |
| Add FGRM tuning to segmentation backbone | Segmentation Backbone · CholecSeg8K LC | DICE: 71.22 -> 74.88 | Improved |
| Add FGRM tuning to segmentation backbone | Segmentation Backbone · ESD dataset | DICE: 84.12 -> 87.23 | Improved |
| Use FGRM instead of MC-dropout (single forward pass) | FGRM · CholecSeg8K LC | Runtime (ms): 0.510 -> 0.052 | Improved |
| Use FGRM instead of MC-dropout (single forward pass) | FGRM · ESD dataset | Runtime (ms): 0.417 -> 0.046 | Improved |
| Use FGRM instead of Deep Ensemble | FGRM · CholecSeg8K LC | ECE: 16.20 -> 9.63 | Improved |
| Add reward maximization, fine-grained update and evidential learning one by one to plain MLE (ablation) | FGRM · LC and ESD | uncertainty quality | Improved |
| Fine-grained parameter update across KL penalty strengths (ablation) | FGRM · LC and ESD | uncertainty quality | Improved |
Closest papers by meaning
- Ctrl-U: Robust Conditional Image Generation via Uncertainty-aware Reward Modeling (ICLR 2025 · similarity 0.87)
- Uncertainty-aware Fine-tuning of Segmentation Foundation Models (NeurIPS 2024 · similarity 0.87)
- Predictive Uncertainty Quantification for Bird's Eye View Segmentation: A Benchmark and Novel Loss Function (ICLR 2025 · similarity 0.86)
- ValUES: A Framework for Systematic Validation of Uncertainty Estimation in Semantic Segmentation (ICLR 2024 · similarity 0.85)
- UncertainSAM: Fast and Efficient Uncertainty Quantification of the Segment Anything Model (ICML 2025 · similarity 0.85)
- SAM-R1: Leveraging SAM for Reward Feedback in Multimodal Segmentation via Reinforcement Learning (NeurIPS 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.
