ChangeDiff

(AAAI-2025) ChangeDiff: A Multi-Temporal Change Detection Data Generator with Flexible Text Prompts via Diffusion Model

This is a pytorch implementation of our paper ChangeDiff. (AAAI-2025)

:speech_balloon: Multi-temporal semantic change synthetic data

It is trained on the sparsely labeled semantic change detection SECOND (Yang et al. 2021) dataset. More sampled synthetic images are available:

:speech_balloon: ChangeDiff Pipeline

:speech_balloon: ChangeDiff Training, sampling.

Please find the corresponding training, sampling and testing scripts under the corresponding files.

Environment Setup

Please follow the below steps:

conda create -n ChangeDiff python=3.8.5
conda activate ChangeDiff
conda install pytorch==1.13.1 torchvision==0.14.1 torchaudio==0.13.1 pytorch-cuda=11.7 -c pytorch -c nvidia
pip install -r requirements.txt

Dataset Setup

If you want to use your own data, please refer to preprocess_data for details.

1. Setup the Second Data dataset

cd train/data
# download Second dataset
# 

Form the second label as follows

train/data/
    label1/
    label2/

Run the preprocessing code:

python preprocess_data/merge_label.py
python preprocess_data/split.py

After this, The users should from the following train/data directory:

train/data/
    coco_gsam_img/
        train/
			metadata.jsonl
            000000000142.jpg
            000000000370.jpg
            ...
    second_layout/
        label1_00001/
            mask_label1_00001_ground.png
            mask_label1_00001_low vegetation.png
            ...
        label1_00011/
            mask_label1_00011_building.png
            mask_label1_00011_ground.png
			mask_label1_00011_tree.png
            ...
        ...

Training

To run T2L, use the following command:

cd train
bash run.sh

The results will be saved under train/results directory.

Sample layout from text

To sample continuous layouts using T2L, use the following command:

cd infer
bash run_rs.sh

The results will be saved under train/results directory.

License

This repository is released under the Apache 2.0 license.

Acknowledge

Some codes are adapted from FreestyleNet, TokenCompose and A2Net. We thank them for their excellent projects.

Citation

If you find this code useful please consider citing

@misc{zang2024ChangeDiff,
      title={ChangeDiff: A Multi-Temporal Change Detection Data Generator with Flexible Text Prompts via Diffusion Model}, 
      author={Qi Zang and Jiayi Yang and Shuang Wang and Dong Zhao and Wenjun Yi and Zhun Zhong},
      year={2024},
      eprint={2412.15541},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2412.15541}, 
}