Entity

Time filter

Source Type

Beijing, China

Wang L.,Beihang University | Jiang K.,Science China Press Co. | Zhou B.,Beihang University | Fu Q.,Beihang University | And 2 more authors.
Proceedings - 2014 International Conference on Virtual Reality and Visualization, ICVRV 2014 | Year: 2015

We introduce a morph able template for dressed human modeling, called Morph able Dressed Human (MDH) template. The template is obtained by learning basic functions from two common types of clothes (long shirts and trousers) and defining a deform-combine manipulation. Our MDH template is parametric, spans variation in both clothes and body/pose, and can generate various dressed human shapes. The template can be morphed to perform dressed human modeling from just a single image through shape fitting and deformation/texture transfer. We demonstrate the effectiveness of our MDH template in single-view dressed human estimation and garment transfer. © 2014 IEEE. Source


Zhang H.,Beihang University | Jiang K.,Science China Press Co. | Zhang Y.,Beihang University | Li Q.,Beihang University | And 2 more authors.
Proceedings - 2014 International Conference on Virtual Reality and Visualization, ICVRV 2014 | Year: 2015

In this paper, we propose a novel deep learning based method for video semantic segmentation. Specially, we utilize 3D convolution neural network (3D CNN) to learn discriminative hierarchical features from spatial-temporal volumes for accurate pixel labelling. The learned features are capable of capturing both appearance and motion information. To align the pixel labels along real object boundaries, as well as maintain local consistency, we further perform graph-cut on a graph constructed on coherent 3d regions, or super-voxels, extracted from input video. Experiments demonstrate that due to the discriminative capability of learned features, our approach can obtain competitive labelling accuracy compared to the state-of-art in absence of sophisticated inference models, even with few training samples. © 2014 IEEE. Source


Tang M.,Beihang University | Tang M.,Science China Press Co. | Yang T.,Beihang University | Zhang Y.,Beihang University
Science China Technological Sciences | Year: 2015

The layer structured zirconium phosphate (ZrPO3) can be intercalated with atoms, molecules, small organic groups and even polymers. The structures and properties of the ZrPO3 intercalation compounds can be deliberately tuned, leading to promising potential applications in many fields. This article provides a brief review on the experimental results of the ZrPO3 intercalation compounds, with the focus on the polymer/α-ZrPO3 nano-composites. The computer simulations of the ZrPO3 intercalation compounds at the atomic level play a significant role in designing and understanding the properties of ZrPO3, and in the promotion of the applications of compounds. © 2015 Science China Press and Springer-Verlag Berlin Heidelberg Source

Discover hidden collaborations