arxivcs.CV2026-07-19
SkyVLaM: Multimodal Large Language Model for UAV Video Understanding in Remote Sensing
Kaiwen Jing, Ruixu Jia, Bingyao Li, Ruizhe Ou, Ming Wu, Chuang Zhang
Recent advances in Multimodal Large Language Models (MLLMs) have significantly improved remote sensing (RS) multimodal understanding. Language-conditioned segmentation is crucial for fine-grained target understanding in Unmanned Aerial Vehicle (UAV) videos. However, this task rem…