Image-Driven Furniture Style for Interactive 3D Scene Modeling
This addresses the challenge for interior designers or users in creating realistic styled spaces more efficiently, though it is incremental as it builds on existing deep learning techniques for style analysis.
The paper tackles the problem of manually selecting style-compatible furniture from large 3D model repositories, which is laborious and time-consuming, by proposing a method that learns furniture style-compatibility from interior scene images using a deep learning network and demonstrates results with an interactive system for modeling style-consistent scenes.
Creating realistic styled spaces is a complex task, which involves design know-how for what furniture pieces go well together. Interior style follows abstract rules involving color, geometry and other visual elements. Following such rules, users manually select similar-style items from large repositories of 3D furniture models, a process which is both laborious and time-consuming. We propose a method for fast-tracking style-similarity tasks, by learning a furniture's style-compatibility from interior scene images. Such images contain more style information than images depicting single furniture. To understand style, we train a deep learning network on a classification task. Based on image embeddings extracted from our network, we measure stylistic compatibility of furniture. We demonstrate our method with several 3D model style-compatibility results, and with an interactive system for modeling style-consistent scenes.