全属性：面向视觉概念个性化的开放词汇属性编码器

摘要

视觉概念个性化旨在仅将特定图像属性（如身份、表情、光照和风格）迁移至未知场景。然而现有方法依赖通用图像编码器的整体嵌入表示，这种表示会纠缠多种视觉因素，导致难以分离单一属性，常引发信息泄露与合成不一致问题。为突破此局限，我们提出Omni-Attribute——首个开放词汇的图像属性编码器，专门用于学习高保真度的属性特定表征。我们的方法协同设计了数据与模型：（一）构建带有正负属性标注的语义关联图像对，显式指导编码器保留或抑制特定信息；（二）采用生成保真度与对比解耦双目标平衡的训练范式。实验表明，所得嵌入表示在开放词汇属性检索、个性化及组合生成任务中效果显著，在多项基准测试中达到最先进性能。

English

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.

全属性：面向视觉概念个性化的开放词汇属性编码器

Omni-Attribute: Open-vocabulary Attribute Encoder for Visual Concept Personalization

摘要

Support