Research map / Detection and segmentation / CNN features
Lightweight Transformer for Multi-Modal Object Detection (Student Abstract)
Yue Cao, Yanshuo Fan, Junchi Bin, Zheng Liu
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 |
|---|---|---|---|
| Use pooling instead of multi-head attention in the fusion operator | PoolFuser · LLVIP | mAP: 0.369 -> 0.384 | Improved |
| Use pooling instead of multi-head attention in the fusion operator | PoolFuser · LLVIP | AP50: 0.780 -> 0.803 | Improved |
| Use pooling instead of multi-head attention in the fusion operator | PoolFuser · LLVIP | GPU Memory: 1940 MB -> 1699 MB | Improved |
| Use pooling instead of multi-head attention in the fusion operator | PoolFuser · LLVIP | Inference time: 0.0712 s -> 0.0490 s | Improved |
| Fuse RGB and IR features with PoolFuser rather than summation | Faster R-CNN · LLVIP | mAP: 0.371 -> 0.384 | Improved |
| Fuse RGB and IR features with PoolFuser rather than summation | Faster R-CNN · LLVIP | AP50: 0.714 -> 0.803 | Improved |
| Fuse RGB and IR features with PoolFuser rather than summation | Faster R-CNN · LLVIP | Inference time: 0.0307 s -> 0.0490 s | Hurt |
| Use two-modal fusion rather than a single modality | Faster R-CNN · LLVIP | mAP: 0.327 -> 0.384 | Improved |
| Use two-modal fusion rather than a single modality | Faster R-CNN · LLVIP | AP50: 0.514 -> 0.803 | Improved |
Closest papers by meaning
- DAMSDet: Dynamic Adaptive Multispectral Detection Transformer with Competitive Query Selection and Adaptive Feature Fusion (ECCV 2024 · similarity 0.86)
- Tri-Modal Fusion Transformers for UAV-based Object Detection (CVPR 2026 · similarity 0.85)
- Efficient Decoder-Free Object Detection with Transformers (ECCV 2022 · similarity 0.85)
- ViDT: An Efficient and Effective Fully Transformer-based Object Detector (ICLR 2022 · similarity 0.84)
- DETRs Beat YOLOs on Real-time Object Detection (CVPR 2024 · similarity 0.82)
- E2E-MFD: Towards End-to-End Synchronous Multimodal Fusion Detection (NeurIPS 2024 · similarity 0.82)
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.
