Research map / Detection and segmentation / Open-set and few-shot detection
Thermal-Det: Language-Guided Cross-Modal Distillation for Open-Vocabulary Thermal Object Detection
Yasiru Ranasinghe, Elim Schenck, Florence Yellin, Shuowen Hu, Christopher Funk, Vishal M. Patel
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 |
|---|---|---|---|
| add scene-level caption loss (ablation) | Thermal-Det · FLIR-Aligned | AP: +0.012 | Improved |
| add scene-level caption loss (ablation) | Thermal-Det · CAMEL | AP: +0.006 | Improved |
| add object-level caption loss (ablation) | Thermal-Det · FLIR-Aligned | AP: +0.028 | Improved |
| add object-level caption loss (ablation) | Thermal-Det · CAMEL | AP: +0.020 | Improved |
| add RGB-to-thermal cross-modal distillation (L_KD) (ablation) | Thermal-Det · FLIR-Aligned | AP: +0.021 | Improved |
| add RGB-to-thermal cross-modal distillation (L_KD) (ablation) | Thermal-Det · CAMEL | AP: +0.012 | Improved |
| add TTAH loss (ablation) | Thermal-Det · FLIR-Aligned | AP: 0.008 AP / 0.019 AP50 / 0.006 AP75 | Improved |
| use confidence-gated subclass selection instead of average pooling (ablation) | TTAH · FLIR-Aligned | AP: 0.234 -> 0.261 | Improved |
| use confidence-gated subclass selection instead of random (ablation) | TTAH · FLIR-Aligned | AP: 0.248 -> 0.261 | Improved |
| use confidence-gated subclass selection instead of average pooling (ablation) | TTAH · CAMEL | AP: 0.559 -> 0.585 | Improved |
| apply thermal adaptation and distillation to MM-GDINO | Thermal-Det · FLIR-Aligned | AP: 0.170 -> 0.234 | Improved |
| apply thermal adaptation and distillation to LLMDet | Thermal-Det · CAMEL | AP: 0.552 -> 0.574 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- LLMDet — LLMDet: Learning Strong Open-Vocabulary Object Detectors under the Supervision of Large Language Models (CVPR 2025) · better on 4 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- ThermalGen: Style-Disentangled Flow-Based Generative Models for RGB-to-Thermal Image Translation (NeurIPS 2025 · similarity 0.86)
- Pseudo Visible Feature Fine-Grained Fusion for Thermal Object Detection (CVPR 2025 · similarity 0.85)
- LLMDet: Learning Strong Open-Vocabulary Object Detectors under the Supervision of Large Language Models (CVPR 2025 · similarity 0.85)
- Alignment-Free RGB-T Salient Object Detection: A Large-Scale Dataset and Progressive Correlation Network (AAAI 2025 · similarity 0.84)
- DetCLIPv3: Towards Versatile Generative Open-vocabulary Object Detection (CVPR 2024 · similarity 0.84)
- Hyperbolic Learning with Synthetic Captions for Open-World Detection (CVPR 2024 · 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.
