Showing 1–3 of 3 results for author: Shui, P

Large-Scale 3D Shape Reconstruction and Segmentation from ShapeNet Core55

Authors: Li Yi, Lin Shao, Manolis Savva, Haibin Huang, Yang Zhou, Qirui Wang, Benjamin Graham, Martin Engelcke, Roman Klokov, Victor Lempitsky, Yuan Gan, Pengyu Wang, Kun Liu, Fenggen Yu, Panpan Shui, Bingyang Hu, Yan Zhang, Yangyan Li, Rui Bu, Mingchao Sun, Wei Wu, Minki Jeong, Jaehoon Choi, Changick Kim, Angom Geetchandra , et al. (25 additional authors not shown)

Abstract: We introduce a large-scale 3D shape understanding benchmark using data and annotation from ShapeNet 3D object database. The benchmark consists of two tasks: part-level segmentation of 3D shapes and 3D reconstruction from single view images. Ten teams have participated in the challenge and the best performing teams have outperformed state-of-the-art approaches on both tasks. A few novel deep learni… ▽ More We introduce a large-scale 3D shape understanding benchmark using data and annotation from ShapeNet 3D object database. The benchmark consists of two tasks: part-level segmentation of 3D shapes and 3D reconstruction from single view images. Ten teams have participated in the challenge and the best performing teams have outperformed state-of-the-art approaches on both tasks. A few novel deep learning architectures have been proposed on various 3D representations on both tasks. We report the techniques used by each team and the corresponding performances. In addition, we summarize the major discoveries from the reported results and possible trends for the future work in the field. △ Less

Submitted 27 October, 2017; v1 submitted 17 October, 2017; originally announced October 2017.

3D Shape Segmentation via Shape Fully Convolutional Networks

Authors: Pengyu Wang, Yuan Gan, Panpan Shui, Fenggen Yu, Yan Zhang, Songle Chen, Zhengxing Sun

Abstract: We desgin a novel fully convolutional network architecture for shapes, denoted by Shape Fully Convolutional Networks (SFCN). 3D shapes are represented as graph structures in the SFCN architecture, based on novel graph convolution and pooling operations, which are similar to convolution and pooling operations used on images. Meanwhile, to build our SFCN architecture in the original image segmentati… ▽ More We desgin a novel fully convolutional network architecture for shapes, denoted by Shape Fully Convolutional Networks (SFCN). 3D shapes are represented as graph structures in the SFCN architecture, based on novel graph convolution and pooling operations, which are similar to convolution and pooling operations used on images. Meanwhile, to build our SFCN architecture in the original image segmentation fully convolutional network (FCN) architecture, we also design and implement a generating operation} with bridging function. This ensures that the convolution and pooling operation we have designed can be successfully applied in the original FCN architecture. In this paper, we also present a new shape segmentation approach based on SFCN. Furthermore, we allow more general and challenging input, such as mixed datasets of different categories of shapes} which can prove the ability of our generalisation. In our approach, SFCNs are trained triangles-to-triangles by using three low-level geometric features as input. Finally, the feature voting-based multi-label graph cuts is adopted to optimise the segmentation results obtained by SFCN prediction. The experiment results show that our method can effectively learn and predict mixed shape datasets of either similar or different characteristics, and achieve excellent segmentation results. △ Less

Submitted 25 May, 2018; v1 submitted 28 February, 2017; originally announced February 2017.

Comments: We update some missing references about intrinsic CNNs (2018.5.24)

Journal ref: Computer & Graphics 70 (2),(2018), p. 128-139

Internal friction and Jahn-Teller effect in the charge-ordered La1-xCaxMnO3 (0.5<x<0.87)

Authors: R. K. Zheng, R. X. Huang, A. N. Tang, G. Li, X. G. Li, J. N. Wei, J. P. Shui, Z. Yao

Abstract: The Jahn-Teller effect in the charge-ordered (CO) state for La1-xCaxMnO3 (0.5<x<0.87) was studied by measuring the low-temperature powder x-ray diffraction, internal friction, and shear modulus. We find that the electron-lattice interaction with the static Jahn-Teller distortion is the strongest near x=0.75 in the CO state. It was particularly observed that a crossover of the Jahn-Teller vibrati… ▽ More The Jahn-Teller effect in the charge-ordered (CO) state for La1-xCaxMnO3 (0.5<x<0.87) was studied by measuring the low-temperature powder x-ray diffraction, internal friction, and shear modulus. We find that the electron-lattice interaction with the static Jahn-Teller distortion is the strongest near x=0.75 in the CO state. It was particularly observed that a crossover of the Jahn-Teller vibration mode from Q2 to Q3 near x=0.75 induces crossovers of the crystal structure from tetragonally compressed to tetragonally elongated orthorhombic, and of the magnetic structure from CE-type to C-type near x=0.75. The experimental results give strong evidence that the Jahn-Teller effect not only plays a key role in stabilizing the CO state, but also determines the magnetic and crystal structures in the CO state for La1-xCaxMnO3. △ Less

Submitted 1 December, 2002; originally announced December 2002.

Comments: 13 pages, 3 figures, PDF

Journal ref: Appl. Phys. Lett. 81(20), 3834 (2002)