Research map / Label-efficient and robust learning / Recognition and long-tail
Deep Embedded Complementary and Interactive Information for Multi-View Classification
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 per-view networks (MvNN vs concatenated mvNN) (ablation) | MvNNcor · Caltech101 | accuracy (%): 60.04 -> 73.24 | Helped |
| add adaptive loss weighting over views (MvNNw vs MvNN) (ablation) | MvNNcor · Caltech101 | accuracy (%): 73.24 -> 73.75 | Helped |
| add adaptive loss weighting over views (MvNNw vs MvNN) (ablation) | MvNNcor · NUSOBJ | accuracy (%): 50.37 -> 48.06 | Hurt |
| add deep interactive network on cross-correlations (MvNNcor vs MvNNw) (ablation) | MvNNcor · NUSOBJ | accuracy (%): 48.06 -> 52.05 | Helped |
| add deep interactive network (MvNNcor vs MvNNw) (ablation) | MvNNcor · AWA | accuracy (%): 45.44 -> 47.69 | Helped |
| batch size 64 vs 32 (ablation) | MvNNcor · Caltech101 | accuracy (%): 74.25 -> 76.00 | Helped |
| interactive network hidden units 200 vs 50 (ablation) | MvNNcor · Caltech101 | accuracy (%): 73.55 -> 76.00 | Helped |
| MvNNcor vs concatenation SVM baseline | MvNNcor · Caltech101 | accuracy (%): 47.90 -> 76.00 | Helped |
| MvNNcor on handcrafted features vs fine-tuned ResNet-101 | MvNNcor · Caltech101 | accuracy (%): 92.43 -> 98.10 | Helped |
Closest papers by meaning
- Rethinking Multi-view Representation Learning via Distilled Disentangling (CVPR 2024 · similarity 0.87)
- MetaViewer: Towards a Unified Multi-View Representation (CVPR 2023 · similarity 0.87)
- Multi-View Deep Network for Cross-View Classification (CVPR 2016 · similarity 0.87)
- Enhanced then Progressive Fusion with View Graph for Multi-View Clustering (CVPR 2025 · similarity 0.84)
- SeqMvRL: A Sequential Fusion Framework for Multi-view Representation Learning (CVPR 2025 · similarity 0.84)
- Trusted Unified Feature-Neighborhood Dynamics for Multi-View Classification (AAAI 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.