Saved in:
Bibliographic Details
Main Authors: Chen, Tsai-Shien, Siarohin, Aliaksandr, Qian, Gordon Guocheng, Wang, Kuan-Chieh Jackson, Nemchinov, Egor, Haji-Ali, Moayed, Guler, Riza Alp, Menapace, Willi, Skorokhodov, Ivan, Kag, Anil, Zhu, Jun-Yan, Tulyakov, Sergey
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.10955
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917451323670528
author Chen, Tsai-Shien
Siarohin, Aliaksandr
Qian, Gordon Guocheng
Wang, Kuan-Chieh Jackson
Nemchinov, Egor
Haji-Ali, Moayed
Guler, Riza Alp
Menapace, Willi
Skorokhodov, Ivan
Kag, Anil
Zhu, Jun-Yan
Tulyakov, Sergey
author_facet Chen, Tsai-Shien
Siarohin, Aliaksandr
Qian, Gordon Guocheng
Wang, Kuan-Chieh Jackson
Nemchinov, Egor
Haji-Ali, Moayed
Guler, Riza Alp
Menapace, Willi
Skorokhodov, Ivan
Kag, Anil
Zhu, Jun-Yan
Tulyakov, Sergey
contents Visual concept personalization aims to transfer only specific image attributes, such as identity, expression, lighting, and style, into unseen contexts. However, existing methods rely on holistic embeddings from general-purpose image encoders, which entangle multiple visual factors and make it difficult to isolate a single attribute. This often leads to information leakage and incoherent synthesis. To address this limitation, we introduce Omni-Attribute, the first open-vocabulary image attribute encoder designed to learn high-fidelity, attribute-specific representations. Our approach jointly designs the data and model: (i) we curate semantically linked image pairs annotated with positive and negative attributes to explicitly teach the encoder what to preserve or suppress; and (ii) we adopt a dual-objective training paradigm that balances generative fidelity with contrastive disentanglement. The resulting embeddings prove effective for open-vocabulary attribute retrieval, personalization, and compositional generation, achieving state-of-the-art performance across multiple benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2512_10955
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Omni-Attribute: Open-vocabulary Attribute Encoder for Visual Concept Personalization
Chen, Tsai-Shien
Siarohin, Aliaksandr
Qian, Gordon Guocheng
Wang, Kuan-Chieh Jackson
Nemchinov, Egor
Haji-Ali, Moayed
Guler, Riza Alp
Menapace, Willi
Skorokhodov, Ivan
Kag, Anil
Zhu, Jun-Yan
Tulyakov, Sergey
Computer Vision and Pattern Recognition
Visual concept personalization aims to transfer only specific image attributes, such as identity, expression, lighting, and style, into unseen contexts. However, existing methods rely on holistic embeddings from general-purpose image encoders, which entangle multiple visual factors and make it difficult to isolate a single attribute. This often leads to information leakage and incoherent synthesis. To address this limitation, we introduce Omni-Attribute, the first open-vocabulary image attribute encoder designed to learn high-fidelity, attribute-specific representations. Our approach jointly designs the data and model: (i) we curate semantically linked image pairs annotated with positive and negative attributes to explicitly teach the encoder what to preserve or suppress; and (ii) we adopt a dual-objective training paradigm that balances generative fidelity with contrastive disentanglement. The resulting embeddings prove effective for open-vocabulary attribute retrieval, personalization, and compositional generation, achieving state-of-the-art performance across multiple benchmarks.
title Omni-Attribute: Open-vocabulary Attribute Encoder for Visual Concept Personalization
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.10955