Research map / Detection and segmentation / CNN features
Multimodal Object Detection via Probabilistic Ensembling
Yi-Ting Chen, Jinghao Shi, Zelin Ye, Christoph Mertz, Deva Ramanan, Shu Kong
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 |
|---|---|---|---|
| replace average score fusion with ProbEn (ablation) | late fusion of RGB+Thermal · KAIST (all-day, reasonable) | LAMR: 19.53 -> 8.62 | Improved |
| replace max-score (NMS) fusion with ProbEn (ablation) | late fusion of RGB+Thermal · KAIST (all-day, reasonable) | LAMR: 10.78 -> 8.50 | Improved |
| use NMS to fuse instead of pooling (ablation) | late fusion · KAIST (all-day, reasonable) | LAMR: 32.68 -> 10.78 | Improved |
| add MidFusion detector as third model to ProbEn (ablation) | ProbEn · KAIST (all-day, reasonable) | LAMR: 8.50 -> 7.66 | Improved |
| swap in off-the-shelf GAFF mid-fusion detector for in-house MidFusion (ablation) | ProbEn3 · KAIST (all-day, reasonable) | LAMR: 7.66 -> 5.14 | Improved |
| fuse boxes with v-avg instead of argmax (ablation) | ProbEn · KAIST (all-day, reasonable) | LAMR: 8.62 -> 8.50 | Improved |
| fuse boxes with avg instead of argmax (ablation) | ProbEn3 (three models) · KAIST (all-day, reasonable) | LAMR: 11.00 -> 7.68 | Improved |
| sum logits instead of averaging logits (ablation) | logit fusion · KAIST (all-day, reasonable) | LAMR: 16.28 -> 8.62 | Improved |
| use learned logistic regression fusion instead of ProbEn (ablation) | late fusion · KAIST (all-day, reasonable) | LAMR: 8.62 -> 9.08 | No clear effect |
| add mid-fusion of RGB and thermal inputs vs thermal only | detector · FLIR (all) | AP: 79.24 -> 80.53 | Improved |
| replace max-score fusion with ProbEn3 on FLIR (ablation) | late fusion · FLIR (all) | AP: 83.14 -> 83.76 | Improved |
| apply ProbEn3 on FLIR vs prior art | ProbEn3 · FLIR (all) | AP: 74.6 -> 84.4 | Improved |
Closest papers by meaning
- Distribution-Aligned Multimodal Fusion for Robust Object Detection (CVPR 2026 · similarity 0.86)
- DiMSOD: A Diffusion-Based Framework for Multi-Modal Salient Object Detection (AAAI 2025 · similarity 0.86)
- E2E-MFD: Towards End-to-End Synchronous Multimodal Fusion Detection (NeurIPS 2024 · similarity 0.86)
- Roboflow100-VL: A Multi-Domain Object Detection Benchmark for Vision-Language Models (NeurIPS 2025 · similarity 0.85)
- Cocoon: Robust Multi-Modal Perception with Uncertainty-Aware Sensor Fusion (ICLR 2025 · similarity 0.85)
- Alignment-Free RGB-T Salient Object Detection: A Large-Scale Dataset and Progressive Correlation Network (AAAI 2025 · 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.
