AccDiffusion: An Accurate Method for Higher-Resolution Image Generation
This addresses a specific issue in image generation for AI applications, but it is incremental as it builds on existing patch-wise methods.
The paper tackles the object repetition problem in patch-wise higher-resolution image generation by proposing AccDiffusion, a method that decouples prompts and uses dilated sampling, resulting in effective reduction of repeated objects and improved performance.
This paper attempts to address the object repetition issue in patch-wise higher-resolution image generation. We propose AccDiffusion, an accurate method for patch-wise higher-resolution image generation without training. An in-depth analysis in this paper reveals an identical text prompt for different patches causes repeated object generation, while no prompt compromises the image details. Therefore, our AccDiffusion, for the first time, proposes to decouple the vanilla image-content-aware prompt into a set of patch-content-aware prompts, each of which serves as a more precise description of an image patch. Besides, AccDiffusion also introduces dilated sampling with window interaction for better global consistency in higher-resolution image generation. Experimental comparison with existing methods demonstrates that our AccDiffusion effectively addresses the issue of repeated object generation and leads to better performance in higher-resolution image generation.