metadata
license: cc-by-4.0
The ChatEarthNet dataset with shortened captions (using LLaVa-v1.6-Mistral7B)
ChatEarthNet: A Global-Scale Image-Text Dataset Empowering Vision-Language Geo-Foundation Models
Yuan, Zhenghang and Xiong, Zhitong and Mou, Lichao and Zhu, Xiao Xiang