• Unified Framework: Supports multiple change detection and captioning tasks.
• Highly Efficient: Uses ~6–13% of the parameters and ~8–34% of the FLOPs compared to SOTA.
• SOTA Performance: Achieves SOTA performance without complex structures, offering an alternative to 2D models.

• [2025.04.04] Change3D has been awarded as a Highlight paper! 🔥🔥🔥

• [2025.03.25] We have released all the training codes for Change3D!

• [2025.02.27] Change3D has been accepted by CVPR 2025! 🎉🎉🎉

We present Change3D, a unified video-based framework for change detection and captioning. Unlike traditional methods that use separate image encoders and multiple change extractors, Change3D treats bi-temporal images as a short video with learnable perception frames. A video encoder enables direct interaction and difference detection, simplifying the architecture. Our approach supports various tasks, including binary change detection (BCD), semantic change detection (SCD), building damage assessment (BDA), and change captioning (CC). Evaluated on eight benchmarks, Change3D outperforms SOTA methods while using only ~6%–13% of the parameters and ~8%–34% of the FLOPs.

Figure 1. Overall architectures of Change3D for Binary Change Detection, Semantic Change Detection, Building Damage Assessment, and Change Captioning.

We conduct extensive experiments on eight public datasets: LEVIR-CD, WHU-CD, CLCD, HRSCD, SECOND, xBD, LEVIR-CC, and DUBAI-CC.

conda create -n Change3D python=3.11.0
conda activate Change3D
pip install torch==2.2.0 torchvision==0.17.0 torchaudio==2.2.0 --index-url https://download.pytorch.org/whl/cu118
pip install -r requirements.txt

Download the X3D-L weight and put it into the root directory.

• For BCD: Download LEVIR-CD, WHU-CD and CLCD datasets. Prepare the dataset into the following structure and crop each image into 256x256 patches.

    ├─Train
        ├─t1          jpg/png (input image of T1)
        ├─t2          jpg/png (input image of T2)
        └─label       jpg/png (binary change mask)
    ├─Val
        ├─t1 
        ├─t2
        └─label
    ├─Test
        ├─t1
        ├─t2
        └─label

• For SCD: Download HRSCD and SECOND datasets. Prepare the dataset into the following structure and crop each image into 256x256 patches.

    ├─Train
        ├─t1          jpg/png  (input image of T1)
        ├─t2          jpg/png  (input image of T2)
        ├─label1      jpg/png  (semantic mask of T1)
        ├─label2      jpg/png  (semantic mask of T2)
        └─change      jpg/png  (binary change mask)
    ...

    ├─Test
        ├─t1
        ├─t2
        ├─label1
        ├─label2
        └─change

• For BDA: Download xBD dataset. Prepare the dataset into the following structure and crop each image into 256x256 patches.

    ├─Train
        ├─t1          jpg/png  (input image of T1)
        ├─t2          jpg/png  (input image of T2)
        ├─label1      jpg/png  (damage localization mask)
        └─label2      jpg/png  (damage level mask)
    ...

    ├─Test
        ├─t1
        ├─t2
        ├─label1
        └─label2

• For CC: Download LEVIR-CC and DUBAI-CC datasets. Then follow the practice introduced in RSICCformer.

Training binary change detection with LEVIR-CD dataset as an example:

python ./scripts/train_BCD.py --dataset LEVIR-CD
                              --file_root path/to/LEVIR-CD
                              --pretrained path/to/X3D_L.pyth
                              --save_dir ./exp
                              --gpu_id 0

Note: The above train script completes the evaluation automatically.

This repository is mainly built upon pytorchvideo and RSICCformer. Thanks for those well-organized codebases.

If you have any issues while using the project, please feel free to contact me: zhuduowang@whu.edu.cn.

Change3D is released under the CC BY-NC-SA 4.0 license.

If you find our work useful, please consider citing our paper:

@InProceedings{Zhu_2025_CVPR,
    author    = {Zhu, Duowang and Huang, Xiaohu and Huang, Haiyan and Zhou, Hao and Shao, Zhenfeng},
    title     = {Change3D: Revisiting Change Detection and Captioning from A Video Modeling Perspective},
    booktitle = {Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR)},
    month     = {June},
    year      = {2025},
    pages     = {24011-24022}
}