Diagnostic Benchmark and Iterative Inpainting for Layout-Guided Image Generation

TL;DR

This paper introduces LayoutBench, a diagnostic benchmark for evaluating layout-guided image generation models, and proposes IterInpaint, a new method that improves out-of-distribution layout generalization through step-by-step inpainting.

Contribution

The paper presents LayoutBench for assessing spatial control skills and introduces IterInpaint, a novel inpainting-based baseline that enhances OOD layout generalization.

Findings

Existing models struggle with OOD layouts, especially at boundaries.

IterInpaint outperforms SOTA models on LayoutBench and LayoutBench-COCO.

Inpainting-based approach improves generalization to unseen layouts.

Abstract

Spatial control is a core capability in controllable image generation. Advancements in layout-guided image generation have shown promising results on in-distribution (ID) datasets with similar spatial configurations. However, it is unclear how these models perform when facing out-of-distribution (OOD) samples with arbitrary, unseen layouts. In this paper, we propose LayoutBench, a diagnostic benchmark for layout-guided image generation that examines four categories of spatial control skills: number, position, size, and shape. We benchmark two recent representative layout-guided image generation methods and observe that the good ID layout control may not generalize well to arbitrary layouts in the wild (e.g., objects at the boundary). Next, we propose IterInpaint, a new baseline that generates foreground and background regions step-by-step via inpainting, demonstrating stronger generalizability than existing models on OOD layouts in LayoutBench. We perform quantitative and qualitative evaluation and fine-grained analysis on the four LayoutBench skills to pinpoint the weaknesses of existing models. We show comprehensive ablation studies on IterInpaint, including training task ratio, crop&paste vs. repaint, and generation order. Lastly, we evaluate the zero-shot performance of different pretrained layout-guided image generation models on LayoutBench-COCO, our new benchmark for OOD layouts with real objects, where our IterInpaint consistently outperforms SOTA baselines in all four splits. Project website: https://layoutbench.github.io