Cloud Classification using Millimeter-wave Cloud Radar Observations and Cloud Images via Visual Language Models
Lists
Only you see this
Abstract
Clouds play a significant role in the Earth’s climate system and meteorological forecasting, and cloud classification is of great importance. In present, millimeter-wave radar and ground-based cloud image are widely employed for cloud classification. However, existing methods only use one type of data. In this paper, we propose to use Visual Language Model (VLM) for cloud classification with both radar data and cloud images. The experimental comparisons verify that our multi-modal method achieves a performance improvement of 10.6% and 8.2% over the best results obtained from the radar-only and image-only modalities, respectively.
BibTeX
@misc{chu2025cloud,
author = {Yu‐Ming Chu and Zeyu Yu and Z. Y. Deng and Zhong Zhang and Shuang Liu and Chaojun Shi},
title = {{Cloud Classification using Millimeter-wave Cloud Radar Observations and Cloud Images via Visual Language Models}},
year = {2025},
doi = {10.1109/aiahpc66801.2025.11289922},
}Loading related content...