SinFusion: Training Diffusion Models on a Single Image or Video

Supplementary Material

Qualitative Comparison to other single-image generative methods

Single Source Image
SinFusion (Ours)
SinGAN [1]
GPNN [2]
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image
Input image
SinFusion image
SinGAN image
GPNN image

Back to Top

Examples of image retargeting

Source Image
Source Image
Retargeted Examples
Source Image
Source Image
Retargeted Examples
Source Image
Source Image
Retargeted Examples
Source Image
Source Image
Retargeted Examples
Source Image
Source Image
Retargeted Examples

Back to Top


Relevant references:
[1] Shaham, T. R., Dekel, T., & Michaeli, T. (2019). Singan: Learning a generative model from a single natural image. In Proceedings of the IEEE/CVF International Conference on Computer Vision (pp. 4570-4580).‏‏
[2] Granot, N., Feinstein, B., Shocher, A., Bagon, S., & Irani, M. (2022). Drop the gan: In defense of patches nearest neighbors as single image generative models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 13460-13469).‏‏‏