Research map / Vision-language and multimodal / Cross-modal learning
Learning Cross-Modal Deep Representations for Robust Pedestrian Detection
Check 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 Sub-Net B initialized with RRN-learned cross-modal representations (ablation) | CMT-CNN · KAIST Reasonable all, day, and night | miss rate: 54.26% -> 49.55%; 52.44% -> 47.30%; 58.97% -> 54.78% | Helped |
| Add Sub-Net B initialized with RRN-learned cross-modal representations rather than ImageNet initialization (ablation) | CMT-CNN · KAIST Reasonable all, day, and night | miss rate: 52.15% -> 49.55%; 50.71% -> 47.30%; 57.65% -> 54.78% | Helped |
| Add RRN-learned transfer to Sub-Net B rather than random initialization (ablation) | CMT-CNN · KAIST Reasonable all, day, and night | miss rate: 56.76% -> 49.55%; 54.83% -> 47.30%; 61.24% -> 54.78% | Helped |
| Add RRN-learned cross-modal representations instead of training Sub-Net B on KAIST RGB (ablation) | CMT-CNN · Caltech-Reasonable | miss rate: 12.51% -> 10.69% | Helped |
| Train Sub-Net B on KAIST RGB rather than use ImageNet initialization alone (ablation) | CMT-CNN-SA-SB · Caltech-Reasonable | miss rate: 13.01% -> 12.51% | Helped |
| Add Sub-Net B with RRN transfer rather than use Sub-Net A only (ablation) | CMT-CNN · Caltech-Reasonable | miss rate: 13.76% -> 10.69% | Helped |
| Vary training batch size from 32 to 256 for the RGB-KAIST Sub-Net B variant (ablation) | CMT-CNN-SA-SB (RGB-KAIST) · Caltech-All and Caltech-Reasonable | miss rate: 65.97%, 65.68%, 65.32%, 65.42%; 13.52%, 13.01%, 12.51%, 12.35% | No clear effect |
Closest papers by meaning
- Robust Pedestrian Detection with Uncertain Modality (AAAI 2026 · similarity 0.88)
- Where, What, Whether: Multi-Modal Learning Meets Pedestrian Detection (CVPR 2020 · similarity 0.88)
- Causal Mode Multiplexer: A Novel Framework for Unbiased Multispectral Pedestrian Detection (CVPR 2024 · similarity 0.85)
- DiMSOD: A Diffusion-Based Framework for Multi-Modal Salient Object Detection (AAAI 2025 · similarity 0.84)
- Implicit Discriminative Knowledge Learning for Visible-Infrared Person Re-Identification (CVPR 2024 · similarity 0.84)
- Shallow-Deep Collaborative Learning for Unsupervised Visible-Infrared Person Re-Identification (CVPR 2024 · similarity 0.83)
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.