Composing People Together: Iterative Pose-Image Generation for Multi-Person Interaction Scenes
This work addresses the challenge of generating complex multi-person interactions for text-to-image models, which is a known bottleneck in the field.
The paper tackles the problem of generating semantically diverse and compositionally accurate multi-person interaction scenes from text, which existing models struggle with. The proposed method introduces a dual pose-image representation and iterative scene construction, achieving substantial improvements in prompt alignment and scene diversity.
Despite recent progress, text-to-image models still struggle to generate semantically diverse and compositionally accurate multi-person interaction scenes, often collapsing to repetitive layouts, stereotypical poses, and poorly grounded interactions. In this work, we bridge this gap by introducing a dual pose-image representation that brings person-centric structural priors into pretrained diffusion transformers. Our model jointly predicts a 2D pose visualization image and its corresponding RGB image, enabling structure and appearance to co-evolve during learning. At its core, a cross-modal alignment scheme binds text, pose, and image representations, ensuring consistent grounding across modalities. Furthermore, we design an iterative scene construction scheme, progressively generating complex multi-human interactions while effectively decomposing the overall generation complexity. Extensive experiments demonstrate that our method substantially improves prompt alignment and scene diversity in multi-person image generation.