Unifying Heterogeneous Multi-Modal Remote Sensing Detection Via Language-Pivoted Pretraining
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917305816973312 |
|---|---|
| author | Li, Yuxuan Chen, Yuming Li, Yunheng Cheng, Ming-Ming Li, Xiang Yang, Jian |
| author_facet | Li, Yuxuan Chen, Yuming Li, Yunheng Cheng, Ming-Ming Li, Xiang Yang, Jian |
| contents | Heterogeneous multi-modal remote sensing object detection aims to accurately detect objects from diverse sensors (e.g., RGB, SAR, Infrared). Existing approaches largely adopt a late alignment paradigm, in which modality alignment and task-specific optimization are entangled during downstream fine-tuning. This tight coupling complicates optimization and often results in unstable training and suboptimal generalization. To address these limitations, we propose BabelRS, a unified language-pivoted pretraining framework that explicitly decouples modality alignment from downstream task learning. BabelRS comprises two key components: Concept-Shared Instruction Aligning (CSIA) and Layerwise Visual-Semantic Annealing (LVSA). CSIA aligns each sensor modality to a shared set of linguistic concepts, using language as a semantic pivot to bridge heterogeneous visual representations. To further mitigate the granularity mismatch between high-level language representations and dense detection objectives, LVSA progressively aggregates multi-scale visual features to provide fine-grained semantic guidance. Extensive experiments demonstrate that BabelRS stabilizes training and consistently outperforms state-of-the-art methods without bells and whistles. Code: https://github.com/zcablii/SM3Det. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_01758 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Unifying Heterogeneous Multi-Modal Remote Sensing Detection Via Language-Pivoted Pretraining Li, Yuxuan Chen, Yuming Li, Yunheng Cheng, Ming-Ming Li, Xiang Yang, Jian Computer Vision and Pattern Recognition Heterogeneous multi-modal remote sensing object detection aims to accurately detect objects from diverse sensors (e.g., RGB, SAR, Infrared). Existing approaches largely adopt a late alignment paradigm, in which modality alignment and task-specific optimization are entangled during downstream fine-tuning. This tight coupling complicates optimization and often results in unstable training and suboptimal generalization. To address these limitations, we propose BabelRS, a unified language-pivoted pretraining framework that explicitly decouples modality alignment from downstream task learning. BabelRS comprises two key components: Concept-Shared Instruction Aligning (CSIA) and Layerwise Visual-Semantic Annealing (LVSA). CSIA aligns each sensor modality to a shared set of linguistic concepts, using language as a semantic pivot to bridge heterogeneous visual representations. To further mitigate the granularity mismatch between high-level language representations and dense detection objectives, LVSA progressively aggregates multi-scale visual features to provide fine-grained semantic guidance. Extensive experiments demonstrate that BabelRS stabilizes training and consistently outperforms state-of-the-art methods without bells and whistles. Code: https://github.com/zcablii/SM3Det. |
| title | Unifying Heterogeneous Multi-Modal Remote Sensing Detection Via Language-Pivoted Pretraining |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.01758 |