A New Volumetric CNN for 3D Object Classification Based on Joint Multiscale Feature and Subvolume Supervised Learning Approaches

Table 1

A comparison between our octree with voxel representation with 32³ volume resolution under the same network (MS-VDCNN-II) on the Modelnet40 dataset.

Representation method		Multiscale	Accuracy (%)
Volume resolution 32^3	Octree	No	91.1
	Voxel	No	90.3
	Octree	Yes	91.4
	Voxel	Yes	90.5