Tackling the Singularities at the Endpoints of Time Intervals in Diffusion Models

Pengze Zhang, Hubery Yin, Chen Li, Xiaohua Xie; Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024, pp. 6945-6954

Abstract


Most diffusion models assume that the reverse process adheres to a Gaussian distribution. However this approximation has not been rigorously validated especially at singularities where t=0 and t=1. Improperly dealing with such singularities leads to an average brightness issue in applications and limits the generation of images with extreme brightness or darkness. We primarily focus on tackling singularities from both theoretical and practical perspectives. Initially we establish the error bounds for the reverse process approximation and showcase its Gaussian characteristics at singularity time steps. Based on this theoretical insight we confirm the singularity at t=1 is conditionally removable while it at t=0 is an inherent property. Upon these significant conclusions we propose a novel plug-and-play method SingDiffusion to address the initial singular time step sampling which not only effectively resolves the average brightness issue for a wide range of diffusion models without extra training efforts but also enhances their generation capability in achieving notable lower FID scores.

Related Material


[pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Zhang_2024_CVPR, author = {Zhang, Pengze and Yin, Hubery and Li, Chen and Xie, Xiaohua}, title = {Tackling the Singularities at the Endpoints of Time Intervals in Diffusion Models}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, month = {June}, year = {2024}, pages = {6945-6954} }