PMMD: A pose-guided multi-view multi-modal diffusion for person generation
This addresses the problem of occlusions, garment style drift, and pose misalignment in person generation for applications like virtual try-on and digital human creation, representing an incremental improvement.
The paper tackles generating consistent human images with controllable pose and appearance, proposing PMMD, a diffusion framework that outperforms baselines in consistency, detail preservation, and controllability on the DeepFashion MultiModal dataset.
Generating consistent human images with controllable pose and appearance is essential for applications in virtual try on, image editing, and digital human creation. Current methods often suffer from occlusions, garment style drift, and pose misalignment. We propose Pose-guided Multi-view Multimodal Diffusion (PMMD), a diffusion framework that synthesizes photorealistic person images conditioned on multi-view references, pose maps, and text prompts. A multimodal encoder jointly models visual views, pose features, and semantic descriptions, which reduces cross modal discrepancy and improves identity fidelity. We further design a ResCVA module to enhance local detail while preserving global structure, and a cross modal fusion module that integrates image semantics with text throughout the denoising pipeline. Experiments on the DeepFashion MultiModal dataset show that PMMD outperforms representative baselines in consistency, detail preservation, and controllability. Project page and code are available at https://github.com/ZANMANGLOOPYE/PMMD.