Observation-Guided Diffusion Probabilistic Models
This work addresses a key bottleneck in diffusion models for image generation, offering an incremental improvement that enhances performance with existing inference techniques.
The authors tackled the tradeoff between quality control and fast sampling in diffusion-based image generation by proposing the observation-guided diffusion probabilistic model (OGDM), which integrates observation process guidance into training to optimize negative log-likelihood, resulting in better denoising networks without extra computational cost.
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.