Research map / Detection and segmentation / CNN features
Modality and Component Aware Feature Fusion For RGB-D Scene 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 modality regularization to L1 for proposal-based FV features (ablation) | Proposal-based FV regression · SUNRGBD | accuracy: 41.0 -> 43.9 | Helped |
| Add component regularization to L1 for proposal-based FV features (ablation) | Proposal-based FV regression · SUNRGBD | accuracy: 41.0 -> 42.7 | Helped |
| Add both modality and component regularization to L1 for proposal-based FV features (ablation) | Proposal-based FV regression · SUNRGBD | accuracy: 41.0 -> 45.1 | Helped |
| Add modality regularization to L1 for full-image CNN features (ablation) | Full-image CNN regression · SUNRGBD | accuracy: 44.9 -> 45.4 | Helped |
| Combine proposal-based FV and full-image regression results (ablation) | Modality and component aware feature fusion · SUNRGBD | accuracy: 45.1 -> 48.1 | Helped |
| Add modality regularization to L1 for proposal-based FV features (ablation) | Proposal-based FV regression · NYU Depth Dataset V2 | accuracy: 53.5 -> 56.7 | Helped |
| Add component regularization to L1 for proposal-based FV features (ablation) | Proposal-based FV regression · NYU Depth Dataset V2 | accuracy: 53.5 -> 55.5 | Helped |
| Add both modality and component regularization to L1 for proposal-based FV features (ablation) | Proposal-based FV regression · NYU Depth Dataset V2 | accuracy: 53.5 -> 59.8 | Helped |
| Add modality regularization to L1 for full-image CNN features (ablation) | Full-image CNN regression · NYU Depth Dataset V2 | accuracy: 58.8 -> 59.1 | Helped |
| Combine proposal-based FV and full-image regression results (ablation) | Modality and component aware feature fusion · NYU Depth Dataset V2 | accuracy: 59.8 -> 63.9 | Helped |
| Add surface-normal features to RGB and HHA (ablation) | Linear SVM on full-image CNN features · SUNRGBD | accuracy: 44.9 -> 45.7 | Helped |
Closest papers by meaning
- Discriminative Multi-Modal Feature Fusion for RGBD Indoor Scene Recognition (CVPR 2016 · similarity 0.90)
- Depth CNNs for RGB-D Scene Recognition: Learning from Scratch Better than Transferring from RGB-CNNs (AAAI 2017 · similarity 0.90)
- Multi-Modal Object Re-identification via Sparse Mixture-of-Experts (ICML 2025 · similarity 0.84)
- Unbiased Faster R-CNN for Single-source Domain Generalized Object Detection (CVPR 2024 · similarity 0.84)
- DiMSOD: A Diffusion-Based Framework for Multi-Modal Salient Object Detection (AAAI 2025 · similarity 0.84)
- Revisiting the Integration of Convolution and Attention for Vision Backbone (NeurIPS 2024 · 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.