Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2604.11415 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914469168283648 |
|---|---|
| author | Xie, Zhenghao Xiao, Jing Wang, Zhenqi Ma, Kexin Liao, Liang Xia, Gui-Song Wang, Mi |
| author_facet | Xie, Zhenghao Xiao, Jing Wang, Zhenqi Ma, Kexin Liao, Liang Xia, Gui-Song Wang, Mi |
| contents | Remote sensing understanding inherently requires multi-resolution observation, since different targets and application tasks demand different levels of spatial detail. While low-resolution (LR) imagery enables efficient global observation, high-resolution (HR) imagery provides critical local details at much higher acquisition cost and limited coverage. This motivates a cross-scale sensing strategy that selectively acquires HR imagery from LR-based global perception to improve task performance under constrained cost. Existing methods for HR sampling methods typically make selection decisions from isolated LR patches, which ignore fine-grained intra-patch importance and cross-patch contextual interactions, leading to fragmented feature representation and suboptimal scene reasoning under sparse HR observations. To address this issue, we formulate cross-scale remote sensing understanding as a unified cost-aware problem that couples fine-grained HR sampling with cross-patch representation prediction, enabling more effective task reasoning with fewer HR observations. Furthermore, we present GL-10M, a large-scale benchmark of 10 million spatially aligned multi-resolution images, enabling systematic evaluation of budget-constrained cross-scale reasoning in remote sensing. Extensive experiments on recognition and retrieval tasks show that our method consistently achieves a superior performance-cost trade-off. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_11415 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Observe Less, Understand More: Cost-aware Cross-scale Observation for Remote Sensing Understanding Xie, Zhenghao Xiao, Jing Wang, Zhenqi Ma, Kexin Liao, Liang Xia, Gui-Song Wang, Mi Computer Vision and Pattern Recognition Remote sensing understanding inherently requires multi-resolution observation, since different targets and application tasks demand different levels of spatial detail. While low-resolution (LR) imagery enables efficient global observation, high-resolution (HR) imagery provides critical local details at much higher acquisition cost and limited coverage. This motivates a cross-scale sensing strategy that selectively acquires HR imagery from LR-based global perception to improve task performance under constrained cost. Existing methods for HR sampling methods typically make selection decisions from isolated LR patches, which ignore fine-grained intra-patch importance and cross-patch contextual interactions, leading to fragmented feature representation and suboptimal scene reasoning under sparse HR observations. To address this issue, we formulate cross-scale remote sensing understanding as a unified cost-aware problem that couples fine-grained HR sampling with cross-patch representation prediction, enabling more effective task reasoning with fewer HR observations. Furthermore, we present GL-10M, a large-scale benchmark of 10 million spatially aligned multi-resolution images, enabling systematic evaluation of budget-constrained cross-scale reasoning in remote sensing. Extensive experiments on recognition and retrieval tasks show that our method consistently achieves a superior performance-cost trade-off. |
| title | Observe Less, Understand More: Cost-aware Cross-scale Observation for Remote Sensing Understanding |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.11415 |