Research map / Detection and segmentation / Open-set and few-shot detection
Visible-Thermal Tiny Object Detection: A Benchmark Dataset and Baselines
Xinyi Ying, Chao Xiao, Wei An, Ruojing Li, Xu He, Boyang Li, Xu Cao, Zhaoxu Li, Yingqian Wang, Mingyuan Hu, Qingyu Xu, Zaiping Lin and 4 more
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 SAFit loss instead of GIoU (ablation) | ATSS · RGBT-Tiny visible | AP: 24.2 -> 24.5 | Improved |
| use SAFit loss instead of NWD (ablation) | ATSS · RGBT-Tiny visible | AP: 24.1 -> 24.5 | Improved |
| use SAFit loss instead of GIoU (ablation) | Sparse RCNN · RGBT-Tiny visible | AP: 19.2 -> 21.4 | Improved |
| use SAFit loss instead of NWD (ablation) | Sparse RCNN · RGBT-Tiny visible | AP: 19.7 -> 21.4 | Improved |
| use SAFit loss instead of plain IoU (ablation) | Sparse RCNN · RGBT-Tiny visible | AP: 8.1 -> 21.4 | Improved |
| sigmoid-weighted SAFit instead of direct switch SAFit-s (ablation) | Sparse RCNN · RGBT-Tiny visible | AP: 19.8 -> 21.4 | Improved |
| sigmoid-weighted SAFit instead of direct switch SAFit-s (ablation) | ATSS · RGBT-Tiny visible | AP: 24.3 -> 24.5 | Improved |
| use stronger components GIoU+NWD (SAFit_g) instead of SAFit (ablation) | Sparse RCNN · RGBT-Tiny visible | AP: 21.4 -> 22.0 | Improved |
| use multimodal fusion instead of replicated visible input (ablation) | ProbEn-middle · RGBT-Tiny visible | AP: 24.7 -> 27.4 | Improved |
| use multimodal fusion instead of replicated visible input (ablation) | UA-CMDet · RGBT-Tiny visible | AP: 18.4 -> 16.1 | Hurt |
Closest papers by meaning
- Alignment-Free RGB-T Salient Object Detection: A Large-Scale Dataset and Progressive Correlation Network (AAAI 2025 · similarity 0.89)
- Towards Large-Scale Small Object Detection: Survey and Benchmarks (TPAMI 2023 · similarity 0.89)
- DVSOD: RGB-D Video Salient Object Detection (NeurIPS 2023 · similarity 0.88)
- XS-VID: A Large-Scale Benchmark for Small Object Detection and Tracking in Videos (TPAMI 2026 · similarity 0.88)
- Towards RAW Object Detection in Diverse Conditions (CVPR 2025 · similarity 0.86)
- DiMSOD: A Diffusion-Based Framework for Multi-Modal Salient Object Detection (AAAI 2025 · similarity 0.85)
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.
