4 Hits in 4.0 sec

SymmetryNet: Learning to Predict Reflectional and Rotational Symmetries of 3D Shapes from Single-View RGB-D Images [article]

Yifei Shi, Junwen Huang, Hongjia Zhang, Xin Xu, Szymon Rusinkiewicz, Kai Xu
2020 arXiv   pre-print
We propose an end-to-end deep neural network which is able to predict both reflectional and rotational symmetries of 3D objects present in the input RGB-D image.  ...  We study the problem of symmetry detection of 3D shapes from single-view RGB-D images, where severely missing data renders geometric detection approach infeasible.  ...  of College of Intelligence Science and Technology, NUDT (2020006, 2020008) and NSF grant IIS-1815070.  ... 
arXiv:2008.00485v3 fatcat:tjm7fpxdgrd4xeicwyis6dgtze

Learning to Detect 3D Reflection Symmetry for Single-View Reconstruction [article]

Yichao Zhou, Shichen Liu, Yi Ma
2020 arXiv   pre-print
3D reconstruction from a single RGB image is a challenging problem in computer vision.  ...  In this work, we focus on object-level 3D reconstruction and present a geometry-based end-to-end deep learning framework that first detects the mirror plane of reflection symmetry that commonly exists  ...  However, unlike multi-view stereopsis, previous single-view methods hardly exploit the geometric constraints between the input RGB image and the resulting 3D shape.  ... 
arXiv:2006.10042v1 fatcat:cqscj5mzjzeepj7yx4bwxsuc7u

Learning to Detect 3D Symmetry From Single-View RGB-D Images With Weak Supervision

Yifei Shi, Xin Xu, Junhua Xi, Xiaochang Hu, Dewen Hu, Kai Xu
In this work, we present a 3D symmetry detection approach to detect symmetry from single-view RGB-D images without using symmetry supervision.  ...  Moreover, to facilitate end-to-end network training and multiple symmetry detection, we introduce a new symmetry parametrization for the learning-based symmetry estimation of both reflectional and rotational  ...  This work was supported in part by NSFC (61825305, 62002379), National Key Research and Development Program of China (2018YFB1305105, 2018AAA0102200), and the Zhejiang Lab's International Talent Fund for  ... 
doi:10.1109/tpami.2022.3186876 pmid:35763472 fatcat:5qlud3qp6jf2nhevczfamhngr4

A Revisit of Shape Editing Techniques: from the Geometric to the Neural Viewpoint [article]

Yu-Jie Yuan, Yu-Kun Lai, Tong Wu, Lin Gao, Ligang Liu
2021 arXiv   pre-print
With increasing availability of 3D shapes on the Internet, data-driven methods were proposed to improve the editing results.  ...  3D shape editing is widely used in a range of applications such as movie production, computer games and computer aided design.  ...  Symmetrynet: Learning to predict reflectional and rotational symmetries of 3d shapes from single-view rgb-d images. arXiv preprint arXiv:2008.00485, 2020. [7] Mitra N J, Wand M, Zhang H, Cohen-Or D, Kim  ... 
arXiv:2103.01694v1 fatcat:lhgswnemnbhvrazl76qhz5rhmy