DeepPano: Deep Panoramic Representation for 3-D Shape Recognition
DeepPano: Deep Panoramic Representation for 3-D Shape Recognition
复制标题
DOI：
10.1109/lsp.2015.2480802
复制
发表时间：
2015-12-01
影响因子：
3.9
通讯作者：
Bai, Xiang
中科院分区：
文献类型：
--
作者：
Shi, Baoguang;Bai, Song;Bai, Xiang
这封信介绍了一种强大的3D形状表示，名为DeepPano，使用深度卷积神经网络（CNN）学习。首先，每个三维形状被转换成一个全景图，即围绕其主轴的圆柱投影。然后，CNN的一个变体被专门设计用于直接从这些视图中学习深度表示。与典型的CNN不同，在卷积层和全连接层之间插入了一个行最大池化层，使学习的表示对围绕主轴的旋转保持不变。我们的方法在两个大规模的3-D模型数据集（ModelNet-10和ModelNet-40）上实现了最先进的检索/分类结果，大大优于典型方法。
This letter introduces a robust representation of 3-D shapes, named DeepPano, learned with deep convolutional neural networks (CNN). Firstly, each 3-D shape is converted into a panoramic view, namely a cylinder projection around its principle axis. Then, a variant of CNN is specifically designed for learning the deep representations directly from such views. Different from typical CNN, a row-wise max-pooling layer is inserted between the convolution and fully-connected layers, making the learned representations invariant to the rotation around a principle axis. Our approach achieves state-of-the-art retrieval/classification results on two large-scale 3-D model datasets (ModelNet-10 and ModelNet-40), outperforming typical methods by a large margin.