Sun, KaifanXiao, JunJiang, HaiyongGünther, TobiasMontazeri, Zahra2025-05-092025-05-092025978-3-03868-269-11017-4656https://doi.org/10.2312/egp.20251018https://diglib.eg.org/handle/10.2312/egp20251018Successful scene arrangement requires ensuring appropriate distances between objects and avoiding excessive overlaps or separations. This work proposes a method for automatically learning spatial relationships between objects in scene arrangement using a differentiable renderer loss. First, objects surrounding a dominant item (e.g., a table in a dining room) are identified and represented as nodes in a polygon that encodes their spatial relations. The difference between the predicted and ground truth polygons is minimized via a rendering loss, which is integrated into the training of a generative diffusion model. This approach continuously optimizes the spatial distribution of objects during generation, ensuring physical consistency and practical usability. Experimental results show a significant reduction in collision rates compared to state-of-the-art methods.Attribution 4.0 International LicenseCCS Concepts: Computing methodologies → Scene generation; Spatial relationship modeling; Differentiable renderingComputing methodologies → Scene generationSpatial relationship modelingDifferentiable rendering10.2312/egp.202510182 pages