HiCo: Hierarchical Controllable Diffusion Model for Layout-to-image Generation

TL;DR

A hierarchical diffusion model is proposed for layout-to-image generation, featuring object seperable conditioning branch structure and key insight is to achieve spatial disentanglement through hierarchical modeling of layouts.

Abstract

The task of layout-to-image generation involves synthesizing images based on the captions of objects and their spatial positions. Existing methods still struggle in complex layout generation, where common bad cases include object missing, inconsistent lighting, conflicting view angles, etc. To effectively address these issues, we propose a \textbf{Hi}erarchical \textbf{Co}ntrollable (HiCo) diffusion model for layout-to-image generation, featuring object seperable conditioning branch structure. Our key insight is to achieve spatial disentanglement through hierarchical modeling of layouts. We use a multi branch structure to represent hierarchy and aggregate them in fusion module. To evaluate the performance of multi-objective controllable layout generation in natural scenes, we introduce the HiCo-7K benchmark, derived from the GRIT-20M dataset and manually cleaned. https://github.com/360CVGroup/HiCo_T2I.

Figures (12)

• Figure 1: HiCo model serves to enhance layout controllability for text-to-image generation, by integrating bounding box condition of different objects hierarchically. The proposed unique conditioning branch structure can produce more harmonious and holistic image with complex layout.
• Figure 2: The generation of CAGchen2024training, GLIGENli2023gligen and HiCo in complex layouts.
• Figure 3: The overall architecture of our approach.
• Figure 4: The visualization of the features on different layers of the HiCo branch and Fuse Net.
• Figure 5: The model fine-tuning technique based on various positions of LoRA. (a) Adding LoRA parameters on UNet. (b) Adding LoRA parameters on HiCo.