Beyond the Clouds: Reliable and Cloud-Aware Spatiotemporal Fusion via Adversarial Regression Wavelets

Lu, Sichen; Li, Mingfei; Jing, Juanjuan; Yu, Junhua; Yang, Lei; Nie, Boyang; Zhou, Jinsong

doi:10.24963/ijcai.2026/163

Beyond the Clouds: Reliable and Cloud-Aware Spatiotemporal Fusion via Adversarial Regression Wavelets

Sichen Lu, Mingfei Li, Juanjuan Jing, Junhua Yu, Lei Yang, Boyang Nie, Jinsong Zhou

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Main Track. Pages 1460-1468.

https://doi.org/10.24963/ijcai.2026/163

Keywords

Computer Vision: Image and video synthesis and generation
Computer Vision: Low-level Vision
Computer Vision: Adversarial learning, adversarial attack and defense methods
Humans and AI: Computational sustainability and human wellbeing

Abstract

Spatiotemporal fusion (STF) bridges the gap between temporal and spatial resolutions in satellite imagery, enabling effective monitoring of Earth's surface dynamics. However, existing methods rely on cloud-free reference images, a constraint that fails in realistic, cloud-prone scenarios. To overcome this, we propose the Cloud-Aware Wavelet Generative Adversarial Network (CLAW-GAN), a novel framework for high-fidelity reconstruction under cloud-contaminated conditions. CLAW-GAN introduces Regression Wavelet Analysis (RWA) to decouple spectral backgrounds from structural details. While a change-aware gated mechanism accounts for land-cover changes, the Frequency-Separated Fusion (FSF) module then independently integrates these components. To ensure visual realism, a multi-scale discriminator operates in the wavelet domain, enforcing consistency across high-frequency subbands to minimize artifacts. Evaluated on the newly introduced Global Cloud-shrouded Agricultural Regions (GCAR) benchmark and the simulated Daxing dataset, CLAW-GAN achieves state-of-the-art performance and demonstrates superior robustness across varying cloud coverage.

Sichen Lu, Mingfei Li, Juanjuan Jing, Junhua Yu, Lei Yang, Boyang Nie, and Jinsong Zhou. 2026. Beyond the Clouds: Reliable and Cloud-Aware Spatiotemporal Fusion via Adversarial Regression Wavelets. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 1460–1468. https://doi.org/10.24963/ijcai.2026/163

@inproceedings{ijcai2026p163,
  title     = {Beyond the Clouds: Reliable and Cloud-Aware Spatiotemporal Fusion via Adversarial Regression Wavelets},
  author    = {Lu, Sichen and Li, Mingfei and Jing, Juanjuan and Yu, Junhua and Yang, Lei and Nie, Boyang and Zhou, Jinsong},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {1460--1468},
  year      = {2026},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2026/163},
  url       = {https://doi.org/10.24963/ijcai.2026/163},
}

TY  - CPAPER
TI  - Beyond the Clouds: Reliable and Cloud-Aware Spatiotemporal Fusion via Adversarial Regression Wavelets
AU  - Lu, Sichen
AU  - Li, Mingfei
AU  - Jing, Juanjuan
AU  - Yu, Junhua
AU  - Yang, Lei
AU  - Nie, Boyang
AU  - Zhou, Jinsong
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 1460
EP  - 1468
DO  - 10.24963/ijcai.2026/163
UR  - https://doi.org/10.24963/ijcai.2026/163
ER  -