Observation-Guided Diffusion Probabilistic Models

We propose a novel diffusion-based image generation method called the observation-guided diffusion probabilistic model (OGDM), which effectively addresses the tradeoff between quality control and fast sampling. Our approach reestablishes the training objective by integrating the guidance of the obse...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Kang, Junoh, Choi, Jinyoung, Choi, Sungik, Han, Bohyung
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 01.04.2024
Subjects
Algorithms
Image processing
Inference
Markov chains
Noise levels
Probabilistic models
Probability theory
Quality control
Training
Online AccessGet full text

Cover

More Information
Summary:We propose a novel diffusion-based image generation method called the observation-guided diffusion probabilistic model (OGDM), which effectively addresses the tradeoff between quality control and fast sampling. Our approach reestablishes the training objective by integrating the guidance of the observation process with the Markov chain in a principled way. This is achieved by introducing an additional loss term derived from the observation based on a conditional discriminator on noise level, which employs a Bernoulli distribution indicating whether its input lies on the (noisy) real manifold or not. This strategy allows us to optimize the more accurate negative log-likelihood induced in the inference stage especially when the number of function evaluations is limited. The proposed training scheme is also advantageous even when incorporated only into the fine-tuning process, and it is compatible with various fast inference strategies since our method yields better denoising networks using the exactly the same inference procedure without incurring extra computational cost. We demonstrate the effectiveness of our training algorithm using diverse inference techniques on strong diffusion model baselines. Our implementation is available at https://github.com/Junoh-Kang/OGDM_edm.
ISSN:2331-8422