Multi-Domain Fusion for UAV Image Super-Resolution Based on Tiny-Transformer

초록

Highlights What are the main findings? A novel multi-domain fusion strategy significantly restores high-frequency textures and repetitive structures in UAV imagery. The proposed model outperforms state-of-the-art methods in both SNR and SSIM metrics, showing significant advantages. What are the implications of the main findings? This work improves the visual quality and usability of UAV imagery through joint multi-domain modeling and cross-domain feature fusion. This work provides an efficient and deployable image enhancement solution for resource-constrained scenarios such as high-altitude urban planning, real-time monitoring, and precision agriculture.Highlights What are the main findings? A novel multi-domain fusion strategy significantly restores high-frequency textures and repetitive structures in UAV imagery. The proposed model outperforms state-of-the-art methods in both SNR and SSIM metrics, showing significant advantages. What are the implications of the main findings? This work improves the visual quality and usability of UAV imagery through joint multi-domain modeling and cross-domain feature fusion. This work provides an efficient and deployable image enhancement solution for resource-constrained scenarios such as high-altitude urban planning, real-time monitoring, and precision agriculture.Abstract Unmanned Aerial Vehicle imagery often suffers from severe spatial detail degradation due to sensor limitations and motion blur, hindering downstream vision tasks. To address this, we propose a lightweight super-resolution framework leveraging a Tiny-Transformer backbone enhanced by a multi-domain feature fusion strategy. Specifically, we jointly model spatial structural semantics and frequency domain texture priors via a cross-domain fusion attention mechanism, enabling coordinated restoration of global consistency and local details. Extensive experiments demonstrate that our method outperforms state-of-the-art approaches on standard benchmarks, achieving significant gains in Peak Signal-to-Noise Ratio and structural similarity index while maintaining low computational cost. Notably, the model exhibits superior robustness in reconstructing high-frequency textures common in aerial scenes. This work provides an efficient, deployable solution for enhancing visual fidelity in resource-constrained applications such as urban planning and precision agriculture.

키워드