Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise
Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise
复制标题
DOI：
10.48550/arxiv.2208.09392
复制
发表时间：
2022-08
期刊：
影响因子：
--
通讯作者：
Arpit Bansal;Eitan Borgnia;Hong-Min Chu;Jie Li;Hamid Kazemi;Furong Huang;Micah Goldblum;Jonas Geiping
中科院分区：
文献类型：
--
作者：
Arpit Bansal;Eitan Borgnia;Hong-Min Chu;Jie Li;Hamid Kazemi;Furong Huang;Micah Goldblum;Jonas Geiping
标准扩散模型涉及图像变换-添加高斯噪声-和图像恢复算子，逆转这种退化。我们观察到，扩散模型的生成行为并不强烈依赖于图像退化的选择，事实上，通过改变这种选择可以构建整个生成模型家族。即使在使用完全确定性降级（例如，模糊、掩蔽等），作为扩散模型基础的训练和测试时更新规则可以很容易地推广以创建生成模型。这些完全确定性模型的成功质疑社区的理解扩散模型，这依赖于噪声的梯度朗之万动力学或变分推理，并铺平了道路，广义扩散模型反转任意过程。我们的代码可在https://github.com/arpitbansal297/Cold-Diffusion-Models上获得
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. Our code is available at https://github.com/arpitbansal297/Cold-Diffusion-Models