Memory Efficient Diffusion Probabilistic Models via Patch-based Generation. (arXiv:2304.07087v1 [cs.CV])

Diffusion probabilistic models have been successful in generating
high-quality and diverse images. However, traditional models, whose input and
output are high-resolution images, suffer from excessive memory requirements,
making them less practical for edge devices. Previous approaches for generative
adversarial networks proposed a patch-based method that uses positional
encoding and global content information. Nevertheless, designing a patch-based
approach for diffusion probabilistic models is non-trivial. In this paper, we
resent a diffusion probabilistic model that generates images on a
patch-by-patch basis. We …

arxiv devices diffusion diverse edge edge devices encoding generative generative adversarial networks global images information making memory networks paper positional encoding practical probabilistic model quality requirements