ScenePhotographer: Object-Oriented Photography for Residential Scenes

Shao-Kui Zhang, Hanxi Zhu, Xuebin Chen, Jinghuan Chen, Zhike Peng, Ziyang Chen, Yong-Liang Yang, Song-Hai Zhang

Abstract: Humans understand digital 3D scenes by observing them from reasonably placed virtual cameras. Selecting camera views is fundamental for 3D scene applications but is typically manual. Existing literature on selecting views is based on regular or polygonal room shapes without focusing on the objects in the scene, resulting in poorly composed views concerning objects. This paper introduces ScenePhotographer, an object-oriented framework for automatic view selection in residential scenes. Potential object-oriented views are yielded by a learning-based method, which clusters objects into groups according to objects' functional and spatial relationships. We propose four criteria to evaluate the views and recommend the best batch, including room information, visibility, composition balance, and line dynamics. Each criterion measures the view according to its corresponding photography rule. Experiments on various room types and layouts demonstrate that our method can generate views focusing on coherent objects while preserving aesthetics, leading to more visually pleasing results.

Primary Subject Area: [Experience] Art and Culture

Secondary Subject Area: [Experience] Multimedia Applications

Relevance To Conference:

This paper presents a method for taking good pictures in 3D residential scenes, leveraging computational interior photography. Firstly, this paper is related to Art and Culture at ACM MM. This paper introduces computational criteria for interior photography, which refers to computational/engineering techniques and artistic/cultural purposes. The criteria include composition balance, line dynamics, etc. Our method is a computational tool to measure views in 3D scenes. Experiments show that our computational photography criteria can aesthetically and informatively guide the views (camera).

Second, this paper is related to Multimedia Applications at ACM MM. Our method generates views for 3D residential scenes, so we also present an application. Interior designers benefit significantly from our method since they no longer need to manually take pictures after designing a room. Our method is successfully integrated into an interior design company for automatic view generation. Furthermore, our method can benefit the metaverse since the digital 3D world needs camera manipulation. We believe our view generation application can improve user experiences related to 3D scenes. Experiments show that our method can generate competitive views with those of professional interior photographers and the existing latest method.

Submission Number: 25