Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise

Standard diffusion models involve an image transform -- adding Gaussian noise -- and an image restoration operator that inverts this degradation. We observe that the generative behavior of diffusion models is not strongly dependent on the choice of image degradation, and in fact, an entire family of generative models can be constructed by varying this choice. Even when using completely deterministic degradations (e. g. , blur, masking, and more), the training and test-time update rules that underlie diffusion models can be easily generalized to create generative models. The success of these fully deterministic models calls into question the community's understanding of diffusion models, which relies on noise in either gradient Langevin dynamics or variational inference and paves the way for generalized diffusion models that invert arbitrary processes.

Authors

• Arpit Bansal
• Eitan Borgnia
• Hong-Min Chu
• Jie Li Shanghai Jiao Tong University Shanghai Jiao Tong University (Wuxi) Blockchain Advanced Research Center
• Hamid Kazemi
• Furong Huang University of Maryland, College Park
• Micah Goldblum
• Jonas Geiping
• Tom Goldstein University of Maryland, College Park

Keywords

No keywords are indexed for this paper.