Manifold Preserving Guided Diffusion
This addresses efficiency and flexibility issues in conditional image generation for low-compute settings, though it appears incremental as it builds on existing diffusion models.
The paper tackles the challenges of cost, generalizability, and task-specific training in conditional image generation by proposing Manifold Preserving Guided Diffusion (MPGD), a training-free framework that uses pretrained diffusion models and off-the-shelf neural networks. It achieves up to 3.8x speed-ups while maintaining high sample quality compared to baselines.
Despite the recent advancements, conditional image generation still faces challenges of cost, generalizability, and the need for task-specific training. In this paper, we propose Manifold Preserving Guided Diffusion (MPGD), a training-free conditional generation framework that leverages pretrained diffusion models and off-the-shelf neural networks with minimal additional inference cost for a broad range of tasks. Specifically, we leverage the manifold hypothesis to refine the guided diffusion steps and introduce a shortcut algorithm in the process. We then propose two methods for on-manifold training-free guidance using pre-trained autoencoders and demonstrate that our shortcut inherently preserves the manifolds when applied to latent diffusion models. Our experiments show that MPGD is efficient and effective for solving a variety of conditional generation applications in low-compute settings, and can consistently offer up to 3.8x speed-ups with the same number of diffusion steps while maintaining high sample quality compared to the baselines.