Research map / Vision-language and multimodal / Cross-modal learning
Towards Versatile Pedestrian Detector with Multisensory-Matching and Multispectral Recalling Memory
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 |
|---|---|---|---|
| Use multisensory-matching contrastive loss and multispectral recalling memory/loss with single-modality inputs | Faster R-CNN with VGG16 · KAIST mixed modality | MR: 20.76 -> 11.39 | Helped |
| Use the proposed detector instead of the strongest listed thermal-based baseline | KAIST thermal input · KAIST | MR: 19.16 -> 15.87 | Helped |
| Use the proposed detector instead of the two-modality baseline | CVC-14 mixed modality · CVC-14 | MR: 28.54 -> 19.88 | Helped |
| Add multisensory-matching contrastive loss to training (ablation) | Proposed detector · KAIST mixed modality | MR: 20.76 -> 15.89 | Helped |
| Add multispectral recalling loss and MSR Memory to training (ablation) | Proposed detector · KAIST mixed modality | MR: 20.76 -> 14.72 | Helped |
| Combine Lmmc and Lmsr (ablation) | Proposed detector · KAIST mixed modality | MR: 20.76 -> 11.39 | Helped |
| Use 25 MSR Memory slots instead of no memory (ablation) | Proposed detector · KAIST mixed modality | MR: 20.76 -> 12.98 | Helped |
| Set MSR Memory slot number to 50 (ablation) | Proposed detector · KAIST mixed modality | MR: 20.76 -> 11.39 | Helped |
| Increase MSR Memory slot number to 100 (ablation) | Proposed detector · KAIST mixed modality | MR: 20.76 -> 12.19 | Helped |
| Increase MSR Memory slot number to 200 (ablation) | Proposed detector · KAIST mixed modality | MR: 20.76 -> 12.58 | Helped |
| Add the proposed method to FPN with ResNet-50 (ablation) | FPN with ResNet-50 · KAIST mixed modality | MR: 18.50 -> 10.68 | Helped |
| Add the proposed method to FPN with ResNet-101 (ablation) | FPN with ResNet-101 · KAIST mixed modality | MR: 17.98 -> 10.32 | Helped |
Closest papers by meaning
- Causal Mode Multiplexer: A Novel Framework for Unbiased Multispectral Pedestrian Detection (CVPR 2024 · similarity 0.86)
- Multispectral Pedestrian Detection with Sparsely Annotated Label (AAAI 2025 · similarity 0.85)
- Implicit Discriminative Knowledge Learning for Visible-Infrared Person Re-Identification (CVPR 2024 · similarity 0.82)
- Bi-directional Adapter for Multimodal Tracking (AAAI 2024 · similarity 0.82)
- Single-Model and Any-Modality for Video Object Tracking (CVPR 2024 · similarity 0.81)
- Collaborating Vision, Depth, and Thermal Signals for Multi-Modal Tracking: Dataset and Algorithm (NeurIPS 2025 · similarity 0.81)
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.