arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2207.09686cs.CV
收藏

对象组合神经隐式曲面

Object-Compositional Neural Implicit Surfaces

Monash University(莫纳什大学);The Chinese University of Hong Kong(香港中文大学);University of Oxford(牛津大学);Nanyang Technological University(南洋理工大学)

Qianyi Wu, Xian Liu, Yuedong Chen, Kejie Li, Chuanxia Zheng, Jianfei Cai, Jianmin Zheng

摘要:

神经隐式表示在多视点图像的新视点合成和高质量三维重建中显示了其有效性。然而,大多数方法专注于整体场景表示,但忽略了其中的单个对象,从而限制了潜在的下游应用。为了学习对象组合表示,一些作品将2D语义图作为训练中的线索,以掌握对象之间的差异。但它们忽略了对象几何信息与实例语义信息之间的紧密联系,导致单个实例的建模不准确。本文提出了一种新的框架,ObjectSDF,建立一个对象组成的神经隐式表示与高保真的三维重建和对象表示。针对传统体渲染流水线的二义性,结合单个物体的符号距离函数(SDF)对场景进行建模,以施加显式的表面约束。区分不同实例的关键是重新审视单个对象的SDF和语义标签之间的强关联。特别地,我们将语义信息转换为对象SDF的函数,并开发了一个统一的和紧凑的表示场景和对象。实验结果表明,ObjectSDF框架在表示整体对象组合场景和单个实例方面的优越性。代码可以在www.example.com上找到

英文摘要:

The neural implicit representation has shown its effectiveness in novel view synthesis and high-quality 3D reconstruction from multi-view images. However, most approaches focus on holistic scene representation yet ignore individual objects inside it, thus limiting potential downstream applications. In order to learn object-compositional representation, a few works incorporate the 2D semantic map as a cue in training to grasp the difference between objects. But they neglect the strong connections between object geometry and instance semantic information, which leads to inaccurate modeling of individual instance. This paper proposes a novel framework, ObjectSDF, to build an object-compositional neural implicit representation with high fidelity in 3D reconstruction and object representation. Observing the ambiguity of conventional volume rendering pipelines, we model the scene by combining the Signed Distance Functions (SDF) of individual object to exert explicit surface constraint. The key in distinguishing different instances is to revisit the strong association between an individual object's SDF and semantic label. Particularly, we convert the semantic information to a function of object SDF and develop a unified and compact representation for scene and objects. Experimental results show the superiority of ObjectSDF framework in representing both the holistic object-compositional scene and the individual instances. Code can be found at https://qianyiwu.github.io/objectsdf/

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑