Search papers, labs, and topics across Lattice.
This paper introduces WaveOp-LiteFM, a lightweight neural operator flow matching framework designed for satellite-to-radar precipitation retrieval, addressing the computational inefficiencies of existing models. By employing a novel spectral-local-wavelet (SLW) block, the framework effectively disentangles precipitation features across multiple frequency regimes, allowing for efficient flow matching in pixel space without sacrificing detail. Experimental results indicate that WaveOp-LiteFM not only achieves state-of-the-art retrieval performance but also significantly reduces computational costs, demonstrating its reliability in large-scale real-world applications such as Typhoon Bavi monitoring.
Achieving state-of-the-art satellite-to-radar precipitation retrieval while slashing computational costs, WaveOp-LiteFM redefines efficiency in meteorological modeling.
Satellite-to-radar (S2R) retrieval refers to estimating ground-based radar precipitation from geostationary satellite observations, enabling precipitation monitoring in regions with limited radar coverage. While recent generative flow matching models have greatly advanced retrieval quality, they face a critical trade-off: pixel-space formulations suffer from the prohibitive computational costs of attention-based U-Net velocity networks, whereas latent-space modeling often sacrifices fine precipitation details or struggles with sparse targets. To address this dilemma, we propose WaveOp-LiteFM, a lightweight neural operator flow matching framework for S2R retrieval. Our approach introduces a novel velocity backbone built upon the spectral-local-wavelet (SLW) block, enabling efficient and stable flow matching in pixel space. Specifically, the SLW block disentangles precipitation features into three distinct frequency regimes: (i) the spectral branch captures large-scale stratiform organization; (ii) the local branch models short-range interactions; and (iii) the wavelet branch enhances sharp structures while suppressing noisy high-frequency responses. Building on this design, an input-adaptive gating mechanism dynamically fuses features from the three functional branches. Furthermore, a skip gate efficiently reintegrates encoder features through additive fusion within the decoder, avoiding the costly channel concatenation used in conventional U-Net architectures. Experiments on the SEVIR and Southeast China datasets show that WaveOp-LiteFM achieves state-of-the-art retrieval performance while substantially reducing computational costs. Beyond benchmark evaluation, large-area inference over China, including a recent Typhoon Bavi case, demonstrates that WaveOp-LiteFM maintains reliable retrieval quality in large-scale real-world scenarios.
