AlignGen: Boosting Personalized Image Generation with Cross-Modality Prior Alignment

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Lin, Yiheng, Zhao, Shifang, Liu, Ting, Qu, Xiaochao, Liu, Luoqi, Zhao, Yao, Wei, Yunchao
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910971368308736
author Lin, Yiheng
Zhao, Shifang
Liu, Ting
Qu, Xiaochao
Liu, Luoqi
Zhao, Yao
Wei, Yunchao
author_facet Lin, Yiheng
Zhao, Shifang
Liu, Ting
Qu, Xiaochao
Liu, Luoqi
Zhao, Yao
Wei, Yunchao
contents Personalized image generation aims to integrate user-provided concepts into text-to-image models, enabling the generation of customized content based on a given prompt. Recent zero-shot approaches, particularly those leveraging diffusion transformers, incorporate reference image information through multi-modal attention mechanism. This integration allows the generated output to be influenced by both the textual prior from the prompt and the visual prior from the reference image. However, we observe that when the prompt and reference image are misaligned, the generated results exhibit a stronger bias toward the textual prior, leading to a significant loss of reference content. To address this issue, we propose AlignGen, a Cross-Modality Prior Alignment mechanism that enhances personalized image generation by: 1) introducing a learnable token to bridge the gap between the textual and visual priors, 2) incorporating a robust training strategy to ensure proper prior alignment, and 3) employing a selective cross-modal attention mask within the multi-modal attention mechanism to further align the priors. Experimental results demonstrate that AlignGen outperforms existing zero-shot methods and even surpasses popular test-time optimization approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2505_21911
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AlignGen: Boosting Personalized Image Generation with Cross-Modality Prior Alignment
Lin, Yiheng
Zhao, Shifang
Liu, Ting
Qu, Xiaochao
Liu, Luoqi
Zhao, Yao
Wei, Yunchao
Computer Vision and Pattern Recognition
Personalized image generation aims to integrate user-provided concepts into text-to-image models, enabling the generation of customized content based on a given prompt. Recent zero-shot approaches, particularly those leveraging diffusion transformers, incorporate reference image information through multi-modal attention mechanism. This integration allows the generated output to be influenced by both the textual prior from the prompt and the visual prior from the reference image. However, we observe that when the prompt and reference image are misaligned, the generated results exhibit a stronger bias toward the textual prior, leading to a significant loss of reference content. To address this issue, we propose AlignGen, a Cross-Modality Prior Alignment mechanism that enhances personalized image generation by: 1) introducing a learnable token to bridge the gap between the textual and visual priors, 2) incorporating a robust training strategy to ensure proper prior alignment, and 3) employing a selective cross-modal attention mask within the multi-modal attention mechanism to further align the priors. Experimental results demonstrate that AlignGen outperforms existing zero-shot methods and even surpasses popular test-time optimization approaches.
title AlignGen: Boosting Personalized Image Generation with Cross-Modality Prior Alignment
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.21911