Git-RSCLIP

Git-RSCLIP is pre-trained on the Git-10M dataset (a global-scale remote sensing image-text pair dataset, consisting of 10 million image-text pairs), available at Github

The paper has been published in IEEE Geoscience and Remote Sensing Magazine: IEEE | ArXiv

News 🔥

✅ 2025.06.01: Git-RSCLIP series downloads exceeded 60,000 times 🔥

Model DownLoad Link

• Large version:[🤗 Huggingface | 🌊 Modelscope]

• Base version: [🤗 Huggingface | 🌊 Modelscope]

Intended uses & limitations

You can use the raw model for tasks like zero-shot image classification and text-image retrieval.

How to use

Use Git-RSCLIP to get image features

```python from PIL import Image import requests from transformers import AutoProcessor, AutoModel import torch

model = AutoModel.frompretrained("lcybuaa/Git-RSCLIP") processor = AutoProcessor.frompretrained("lcybuaa/Git-RSCLIP")

url = "https://github.com/Chen-Yang-Liu/PromptCC/blob/main/Example/B/train_000051.png?raw=true" image = Image.open(requests.get(url, stream=True).raw)

inputs = processor(images=image, return_tensors="pt")

with torch.nograd(): imagefeatures = model.getimagefeatures(**inputs) ```

zero-shot image classification:

```python from PIL import Image import requests from transformers import AutoProcessor, AutoModel import torch

model = AutoModel.frompretrained("lcybuaa/Git-RSCLIP") processor = AutoProcessor.frompretrained("lcybuaa/Git-RSCLIP")

url = "https://github.com/Chen-Yang-Liu/PromptCC/blob/main/Example/B/train_000051.png?raw=true" image = Image.open(requests.get(url, stream=True).raw)

texts = ["a remote sensing image of river", "a remote sensing image of houses and roads"] inputs = processor(text=texts, images=image, padding="maxlength", returntensors="pt")

with torch.no_grad(): outputs = model(**inputs)

logitsperimage = outputs.logitsperimage probs = torch.sigmoid(logitsperimage) # these are the probabilities top5indices = torch.argsort(probs, descending=True)[:, :5].cpu().numpy() top1indices = top5indices[:, 0] print(f"the image 0 is '{top1indices[0]}'") ```

For more code examples, we refer to the documentation.

Training procedure

Training data

Git-RSCLIP is pre-trained on the Git-10M dataset (a global-scale remote sensing image-text pair dataset, consisting of 10 million image-text pairs) (Liu et al., 2024).

Preprocessing

Images are resized/rescaled to the same resolution (256x256) and normalized across the RGB channels with mean (0.5, 0.5, 0.5) and standard deviation (0.5, 0.5, 0.5).

Texts are tokenized and padded to the same length (64 tokens).

Evaluation results

Evaluation of Git-RSCLIP compared to other CLIP is shown below (taken from the paper).

BibTeX entry and citation info

bibtex @ARTICLE{10988859, author={Liu, Chenyang and Chen, Keyan and Zhao, Rui and Zou, Zhengxia and Shi, Zhenwei}, journal={IEEE Geoscience and Remote Sensing Magazine}, title={Text2Earth: Unlocking text-driven remote sensing image generation with a global-scale dataset and a foundation model}, year={2025}, volume={}, number={}, pages={2-23}, doi={10.1109/MGRS.2025.3560455}}