RS5M
RS5M copied to clipboard
RS5M: a large-scale vision language dataset for remote sensing
fmow/train/airport/airport_0/airport_0_0_msrgb.jpg,fmow,fmow_airport_0_0_msrgb.jpg,"above the Winter landscapes of Etimesgut, Turkey, the view is transfixed by airport at its center and center-left blocks. with sharpness ensured by a ground sample distance of 2.45 meters,...
Hello: Thank you for sharing such great dataset and model. But I have a question, I noticed in your paper the method for generating captioins: captions can be employed by...
Great paper and work! I have not found the fine-tuning or training code in the repository. Could you release it? Thanks!
Hello, could you provide the checkpoint fine-tuned solely on RSITMD or RSICD? not RET-2. I cannot reproduce the result reported in the paper. The reproduced result is far worse than...
Thank you very much for your outstanding work! I have a question that I haven't quite understood. When fine-tuning your RS5M model on RSICD or RSITMD using the methods outlined...
感觉作者大大的工作,我在复现GeoRSCLIP遇到了些问题,我 Full-fine-tuning open_clip(RS5M) + rsicd training set 结果在34附近,与论文结果38差很多,佬能提供一些实现细节嘛,期待你的回复。