Make a Cheap Scaling: A Self-Cascade Diffusion Model for Higher-Resolution Adaptation

TL;DR

This paper introduces a self-cascade diffusion model that efficiently adapts pre-trained low-resolution models to generate higher-resolution images and videos, significantly reducing training time and resource requirements.

Contribution

The proposed model enables rapid, tuning-free adaptation to higher resolutions using knowledge transfer and multi-scale upsamplers, with minimal additional training.

Findings

Achieves 5x faster training speed compared to full fine-tuning

Requires only 0.002M tuning parameters

Can adapt to higher resolutions with just 10k training steps

Abstract

Diffusion models have proven to be highly effective in image and video generation; however, they encounter challenges in the correct composition of objects when generating images of varying sizes due to single-scale training data. Adapting large pre-trained diffusion models to higher resolution demands substantial computational and optimization resources, yet achieving generation capabilities comparable to low-resolution models remains challenging. This paper proposes a novel self-cascade diffusion model that leverages the knowledge gained from a well-trained low-resolution image/video generation model, enabling rapid adaptation to higher-resolution generation. Building on this, we employ the pivot replacement strategy to facilitate a tuning-free version by progressively leveraging reliable semantic guidance derived from the low-resolution model. We further propose to integrate a sequence of learnable multi-scale upsampler modules for a tuning version capable of efficiently learning structural details at a new scale from a small amount of newly acquired high-resolution training data. Compared to full fine-tuning, our approach achieves a $5\times$ training speed-up and requires only 0.002M tuning parameters. Extensive experiments demonstrate that our approach can quickly adapt to higher-resolution image and video synthesis by fine-tuning for just $10k$ steps, with virtually no additional inference time.