PortraitCraft: A Benchmark for Portrait Composition Understanding and Generation

TL;DR

PortraitCraft introduces a comprehensive benchmark dataset and evaluation framework for portrait composition understanding and generation, enabling systematic research in structured portrait analysis and controllable image synthesis.

Contribution

The paper presents a new large-scale, multi-level annotated portrait dataset and defines two benchmark tasks for understanding and generating portraits with explicit compositional constraints.

Findings

Baseline models demonstrate the feasibility of composition understanding and generation tasks.

The dataset enables fine-grained attribute reasoning and interpretability in portrait analysis.

Standardized protocols facilitate future research in portrait aesthetics and controllable synthesis.

Abstract

Portrait composition plays a central role in portrait aesthetics and visual communication, yet existing datasets and benchmarks mainly focus on coarse aesthetic scoring, generic image aesthetics, or unconstrained portrait generation. This limits systematic research on structured portrait composition analysis and controllable portrait generation under explicit composition requirements. In this paper, we introduce PortraitCraft, a unified benchmark for portrait composition understanding and generation. PortraitCraft is built on a dataset of approximately 50,000 curated real portrait images with structured multi-level supervision, including global composition scores, annotations over 13 composition attributes, attribute-level explanation texts, visual question answering pairs, and composition-oriented textual descriptions for generation. Based on this dataset, we establish two complementary benchmark tasks for composition understanding and composition-aware generation within a unified framework. The first evaluates portrait composition understanding through score prediction, fine-grained attribute reasoning, and image-grounded visual question answering, while the second evaluates portrait generation from structured composition descriptions under explicit composition constraints. We further define standardized evaluation protocols and provide reference baseline results with representative multimodal models. PortraitCraft provides a comprehensive benchmark for future research on fine-grained portrait understanding, interpretable aesthetic assessment, and controllable portrait generation.