Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Huang, Siteng, Gong, Biao, Feng, Yutong, Chen, Xi, Fu, Yuqian, Liu, Yu, Wang, Donglin
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2311.15841
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917662375804928
author Huang, Siteng
Gong, Biao
Feng, Yutong
Chen, Xi
Fu, Yuqian
Liu, Yu
Wang, Donglin
author_facet Huang, Siteng
Gong, Biao
Feng, Yutong
Chen, Xi
Fu, Yuqian
Liu, Yu
Wang, Donglin
contents This study focuses on a novel task in text-to-image (T2I) generation, namely action customization. The objective of this task is to learn the co-existing action from limited data and generalize it to unseen humans or even animals. Experimental results show that existing subject-driven customization methods fail to learn the representative characteristics of actions and struggle in decoupling actions from context features, including appearance. To overcome the preference for low-level features and the entanglement of high-level features, we propose an inversion-based method Action-Disentangled Identifier (ADI) to learn action-specific identifiers from the exemplar images. ADI first expands the semantic conditioning space by introducing layer-wise identifier tokens, thereby increasing the representational richness while distributing the inversion across different features. Then, to block the inversion of action-agnostic features, ADI extracts the gradient invariance from the constructed sample triples and masks the updates of irrelevant channels. To comprehensively evaluate the task, we present an ActionBench that includes a variety of actions, each accompanied by meticulously selected samples. Both quantitative and qualitative results show that our ADI outperforms existing baselines in action-customized T2I generation. Our project page is at https://adi-t2i.github.io/ADI.
format Preprint
id arxiv_https___arxiv_org_abs_2311_15841
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Learning Disentangled Identifiers for Action-Customized Text-to-Image Generation
Huang, Siteng
Gong, Biao
Feng, Yutong
Chen, Xi
Fu, Yuqian
Liu, Yu
Wang, Donglin
Computer Vision and Pattern Recognition
This study focuses on a novel task in text-to-image (T2I) generation, namely action customization. The objective of this task is to learn the co-existing action from limited data and generalize it to unseen humans or even animals. Experimental results show that existing subject-driven customization methods fail to learn the representative characteristics of actions and struggle in decoupling actions from context features, including appearance. To overcome the preference for low-level features and the entanglement of high-level features, we propose an inversion-based method Action-Disentangled Identifier (ADI) to learn action-specific identifiers from the exemplar images. ADI first expands the semantic conditioning space by introducing layer-wise identifier tokens, thereby increasing the representational richness while distributing the inversion across different features. Then, to block the inversion of action-agnostic features, ADI extracts the gradient invariance from the constructed sample triples and masks the updates of irrelevant channels. To comprehensively evaluate the task, we present an ActionBench that includes a variety of actions, each accompanied by meticulously selected samples. Both quantitative and qualitative results show that our ADI outperforms existing baselines in action-customized T2I generation. Our project page is at https://adi-t2i.github.io/ADI.
title Learning Disentangled Identifiers for Action-Customized Text-to-Image Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2311.15841