Research map / Label-efficient and robust learning / Recognition and long-tail
Trusted Multi-View Classification With Dynamic Evidential Fusion
Zongbo Han, Changqing Zhang, Huazhu Fu, Joey Tianyi Zhou
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 |
|---|---|---|---|
| add a pseudo-view (concatenated features) with its own opinion (ablation) | TMC · Scene15 | accuracy (%): 67.74 -> 71.61 | Improved |
| add a pseudo-view (concatenated features) (ablation) | TMC · HMDB | accuracy (%): 65.26 -> 69.43 | Improved |
| add a pseudo-view (concatenated features) (ablation) | TMC · PIE | accuracy (%): 91.99 -> 93.75 | Improved |
| add a pseudo-view (concatenated features) (ablation) | TMC · Caltech101 | accuracy (%): 92.93 -> 92.83 | No clear effect |
| add a pseudo-view (concatenated backbone outputs) (ablation) | TMC · SUN RGB-D (fusion) | accuracy (%): 60.7 -> 61.3 | Improved |
| add a pseudo-view (concatenated backbone outputs) (ablation) | TMC · SUN RGB-D (RGB only) | accuracy (%): 57.1 -> 56.2 | Hurt |
| add a pseudo-view (concatenated backbone outputs) (ablation) | TMC · NYUD Depth V2 (fusion) | accuracy (%): 70.0 -> 72.5 | Improved |
| use evidence-level uncertainty fusion instead of CCA plus linear SVM | TMC · Scene15 | accuracy (%): 55.77 -> 67.74 | Improved |
| use stronger correlation maximization (DCCA instead of CCA) | CCA baseline · Scene15 | accuracy (%): 55.77 -> 54.85 | Hurt |
| fuse views with uncertainty-based fusion instead of EDL on feature-concatenated input | EDL · UMPC-FOOD101 | accuracy (%): 90.50 -> 91.30 | Improved |
| use feature fusion of image and text instead of best single view | EDL · UMPC-FOOD101 | accuracy (%): 87.40 -> 90.50 | Improved |
| discard samples with uncertainty greater than the largest belief mass | ETMC · SUN RGB-D | misclassification rate: 0.387 -> 0.187 | Improved |
Closest papers by meaning
- Enhancing Testing-Time Robustness for Trusted Multi-View Classification in the Wild (CVPR 2025 · similarity 0.92)
- Trusted Multi-View Classification with Expert Knowledge Constraints (ICML 2025 · similarity 0.89)
- Enhancing Multi-View Classification Reliability with Adaptive Rejection (AAAI 2025 · similarity 0.89)
- Navigating Conflicting Views: Harnessing Trust for Learning (ICML 2025 · similarity 0.89)
- Deep Fuzzy Multi-view Learning for Reliable Classification (ICML 2025 · similarity 0.88)
- Trusted Unified Feature-Neighborhood Dynamics for Multi-View Classification (AAAI 2025 · similarity 0.88)
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.
