Abstract

Diffusion models have recently achieved remarkable progress in generating realistic images. However, challenges remain in accurately understanding and synthesizing the layout requirements in the textual prompts. To align the generated image with layout instructions, we present a training-free layout calibration system SimM that intervenes in the generative process on the fly during inference time. Specifically, following a "check-locate-rectify" pipeline, the system first analyses the prompt to generate the target layout and compares it with the intermediate outputs to automatically detect errors. Then, by moving the located activations and making intra- and inter-map adjustments, the rectification process can be performed with negligible computational overhead. To evaluate SimM over a range of layout requirements, we present a benchmark SimMBench that compensates for the lack of superlative spatial relations in existing datasets. And both quantitative and qualitative results demonstrate the effectiveness of the proposed SimM in calibrating the layout inconsistencies.

Samples

Examples of multi-resolution image generated by SimM.

case

Given only the input textual prompt, SimM can autonomously detect and rectify the layout inconsistencies across various position requirements (a-d), object quantities (e-g), and resolutions (h-i).

teaser

BibTeX

@article{SimM2024,
    title={Check, Locate, Rectify: A Training-Free Layout Calibration System for Text-to-Image Generation}, 
    author={Biao Gong and Siteng Huang and Yutong Feng and Shiwei Zhang and Yuyuan Li and Yu Liu},
    year={2023},
    eprint={2311.15773},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}
Ranni SimM ADI