HiPropmt : Tuning-free Higher-Resolution Generation with Hierarchical MLLM Prompts

Xinyu Liu1, Yingqing He1, Lanqing Guo2, Xiang Li3, Bu Jin4, Peng Li1, Yan Li1, Chi-Min Chan1, Qifeng Chen1, Wei Xue1, Wenhan Luo1, Qifeng Liu1, Yike Guo1
1Hong Kong University of Science and Technology, 2Nanyang Technological University, 3Tsinghua University, 4University of Chinese Academy of Sciences
Choose a thumbnail image at the bottom of the page to explore details. You can zoom in by mdi-gesture-pinch pinching or mdi-swap-vertical scrolling and pan by mdi-gesture-swipe-horizontal mdi-mouse dragging.
{{viewport.label}}
Current zoom: 100% Select background color
open {{tid - 1}}.png

Detailed Comparison

Abstract

The potential for higher-resolution image generation using pretrained diffusion models is immense, yet these models often struggle with issues of object repetition and structural artifacts especially when scaling to 4K resolution and higher. We figure out that the problem is caused by that, a single prompt for the generation of multiple scales provides insufficient efficacy. In response, we propose HiPrompt, a new tuning-free solution that tackles the above problems by introducing hierarchical prompts. The hierarchical prompts offer both global and local guidance. Specifically, the global guidance comes from the user input that describes the overall content, while the local guidance utilizes patch-wise descriptions from MLLMs to elaborately guide the regional structure and texture generation. Furthermore, during the inverse denoising process, the generated noise is decomposed into low- and high-frequency spatial components. These components are conditioned on multiple prompt levels, including detailed patch-wise descriptions and broader image-level prompts, facilitating prompt-guided denoising under hierarchical semantic guidance. It further allows the generation to focus more on local spatial regions and ensures the generated images maintain coherent local and global semantics, structures, and textures with high definition. Extensive experiments demonstrate that HiPrompt outperforms state-of-the-art works in higher-resolution image generation, significantly reducing object repetition and enhancing structural quality.

HiPrompt

Given a low-resolution image, MLLMs are employed to generate dense local descriptions for each overlapping local patch. To enhance the quality of these detailed prompts, we utilize N-grams (n = 1) refinement to filter out irrelevant noise. Subsequently, HiPrompt decomposes the noisy image into low- and high-spatial frequency components using low-pass and high-pass Gaussian filters. These components are denoised in parallel, conditioned on the hierarchical prompts, and then summarized into final estimation during the inverse denoising process.

Qualitative Comparision Results


The red boxes highlight the repeated object problem, while the yellow boxes denote areas with blurred and unreasonable structures.

"A professional photograph of an astronaut riding a horse."

"A cute and adorable fluffy puppy wearing a witch hat in a halloween autumn evening forest, falling autumn leaves, brown acorns on the ground, halloween pumpkins spiderwebs, bats, a witch’s broom."


More Samples at Various Resolutions