Remote Sensing ChatGPT(RS-ChatGPT) is an open source tool for solving remote sensing tasks with ChatGPT in an interactive way.ChatGPT acts as an expert to response to users' linguistic resquests based on the input remote sensing image. It supports various interpretation tasks that are trained on remote sensing datasets. To help ChatGPT better understand remote sensing knowledge, remote sensing image captioning is set as cue to help ChatGPT understand remote sensing image.We will keep updating RS-CHatGPT~ Please comments with Issues or send me a email if you have any suggestions! Thanks in advance!
Generally, RS-ChatGPT includes four steps in implementation:
- Prompt Template Generation
- Task Planning
- Task Execution
- Response Genration
Our article is now available in arXiv! See you in IGARSS204, Greece!
My personal website is also avalable:www.poleguo98.top
- TODO1: An online website is comming! Let's use RS-ChatGPT Online!
- TODO2: What about asking RS-CHatGPT to choose different models within the given task?
- TODO3: Put all the models available in Huggingface
2024.01.30:
- Remote Sensing GPT is updated, with bugs fixed, more readable, and more scalable~
2023.09.19:
- Remote Sensing GPT now supports GPT-4 and multi-round chating! An interactive interface is now available(see interface.py)~
2023.08.23:
- Initial release
Please Refer to requirements.txt
->Clone the repository:git clone https://github.com/HaonanGuo/Remote-Sensing-ChatGPT
->Download the below models and place them in the checkpoints folder
->Run RSChatGPT-shell.py
Function | Description | Method | Pretrain Dataset | Model Weights |
---|---|---|---|---|
Image Captioning | Describe the remote sensing image | BLIP | BLIP Dataset | weight(github) |
Scene Classification | Classify the type of scene | ResNet | AID Dataset | weight(Google) weight(Baidu) |
Object Detection | Detect RS object from image | YOLO v5 | DOTA | weight(Google) weight(Baidu) |
Instance Segmentation | Extract Instance Mask of certain object | SwinTransformer+UperNet | iSAID | weight(Google) weight(Baidu) |
Landuse Classification | Extract Pixel-wise Landuse Classification | HRNet | LoveDA | weight(Google) weight(Baidu) |
Object Counting | Count the number of certain object in an image | YOLO v5 | DOTA | Same as Object Detection |
Edge Detection | Extract edge of remote sensing image | Canny | None | None |
More funtions to be updated~
Please cite the repo if you use the data or code in this repo.
@article{RS ChatGPT,
title = {Remote Sensing ChatGPT: Solving Remote Sensing Tasks with ChatGPT and Visual Models},
shorttitle = {Remote Sensing ChatGPT},
doi = {10.48550/ARXIV.2401.09083},
author = {Guo, Haonan and Su, Xin and Wu, Chen and Du, Bo and Zhang, Liangpei and Li, Deren},
year = {2024},
}
Remote Sensing ChatGPT is an open source project that welcome any contribution and feedback. Please contact us with: [email protected]