BiliSakura/ddpm-cd-levir-50-100

Change detection pipeline — DDPM-CD fine-tuned on LEVIR. Uses SR3 UNet backbone + CD head with timesteps [50, 100].

Model Details

• Dataset: LEVIR
• Timesteps: 50, 100
• Architecture: DDPM (SR3 UNet) + cd_head_v2

Usage

from diffusers import DiffusionPipeline

pipe = DiffusionPipeline.from_pretrained(
    "BiliSakura/ddpm-cd-levir-50-100",
    custom_pipeline="pipeline",
    trust_remote_code=True,
)
# cd_head loads automatically on first __call__
# Images in [-1, 1], shape (B, 3, H, W)
change_map = pipe(image_A, image_B, timesteps=[50, 100])
pred = change_map.argmax(1)  # (B, H, W), 0=no-change, 1=change

Citation

@misc{bandara2024ddpmcdv3,
    title={DDPM-CD: Denoising Diffusion Probabilistic Models as Feature Extractors for Change Detection},
    author={Wele Gedara Chaminda Bandara and Nithin Gopalakrishnan Nair and Vishal M. Patel},
    year={2024},
    eprint={2206.11892},
    archivePrefix={arXiv},
    primaryClass={cs.CV},
}