PairHuman: A High-Fidelity Photographic Dataset for Customized Dual-Person Generation

2025-11-20

AI Summary

Existing dual-portrait generation research is hindered by the absence of high-quality, customizable benchmark datasets. To address this, we introduce DP-Portrait—the first large-scale, photorealistic dual-person image benchmark—featuring diverse human attributes, precise facial landmarks, spatial pose annotations, and fine-grained semantic scene labels. We further propose DHumanDiff, a dedicated diffusion-based generative model that jointly optimizes facial identity preservation and semantically controllable scene synthesis. Extensive experiments demonstrate that DP-Portrait significantly improves visual fidelity, personalization accuracy, and human aesthetic preference scores in generated dual portraits. Both the dataset and baseline models are publicly released to foster standardized advancement in dual-portrait generation and editing.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Humans and AI: Intelligent User Interfaces
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Personalized dual-person portrait customization has considerable potential applications, such as preserving emotional memories and facilitating wedding photography planning. However, the absence of a benchmark dataset hinders the pursuit of high-quality customization in dual-person portrait generation. In this paper, we propose the PairHuman dataset, which is the first large-scale benchmark dataset specifically designed for generating dual-person portraits that meet high photographic standards. The PairHuman dataset contains more than 100K images that capture a variety of scenes, attire, and dual-person interactions, along with rich metadata, including detailed image descriptions, person localization, human keypoints, and attribute tags. We also introduce DHumanDiff, which is a baseline specifically crafted for dual-person portrait generation that features enhanced facial consistency and simultaneously balances in personalized person generation and semantic-driven scene creation. Finally, the experimental results demonstrate that our dataset and method produce highly customized portraits with superior visual quality that are tailored to human preferences. Our dataset is publicly available at https://github.com/annaoooo/PairHuman.

Problem

• Lack of benchmark dataset for high-quality dual-person portrait generation
• Need for enhanced facial consistency in personalized dual-person generation
• Balancing personalized person generation with semantic-driven scene creation

Innovation

• Developed large-scale photographic dataset for dual-person generation
• Introduced baseline model enhancing facial consistency and personalization
• Balanced personalized generation with semantic-driven scene creation