Y. Yoshiyasu, E. Yoshida, S. Pirk, and L. J. Guibas, 3D Convolutional Neural Networks by Modal Fusion, IEEE International Conference on Image Processing (ICIP), 2017.


We propose multi-view and volumetric convolutional neural networks (ConvNets) for 3D shape recognition, which combines surface normal and height fields to capture local geometry and physical size of an object. This strategy helps distinguishing between objects with similar geometries but different sizes. This is especially useful for enhancing volumetric ConvNets and classifying 3D scans with insufficient surface details. Experimental results on CAD and real-world scan datasets showed that our technique outperforms previous approaches.


  title={3D convolutional neural networks by modal fusion},
  author={Yusuke Yoshiyasu and Eiichi Yoshida and S{"o}ren Pirk and Leonidas J. Guibas},
  journal={2017 IEEE International Conference on Image Processing (ICIP)},