3D Object Classification via Spherical Projections
Authors:
- Zhangjie CaoZCZhangjie Cao
- Qixing HuangQHQixing Huang
- Ramani KarthikRKRamani Karthik
- 1 October 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 566-574
Abstract
In this paper, we introduce a new method for classifying 3D objects. Our main idea is to project a 3D object onto a spherical domain centered around its barycenter and develop neural network to classify the spherical projection. We introduce two complementary projections. The first captures depth variations of a 3D object, and the second captures contour-information viewed from different angles. Spherical projections combine key advantages of two main-stream 3D classification methods: image-based and 3D-based. Specifically, spherical projections are locally planar, allowing us to use massive image datasets (e.g, ImageNet) for pre-training. Also spherical projections are similar to voxel-based methods, as they encode complete information of a 3D object in a single neural network capturing dependencies across different views. Our novel network design can fully utilize these advantages. Experimental results on ModelNet40 and ShapeNetCore show that our method is superior to prior methods.Keywords
All Related Versions
This publication has 15 references indexed in Scilit:
- OctNetFusion: Learning Depth Fusion from Data2017
- Beam search for learning a deep Convolutional Neural Network of 3D shapes2016
- PointNet: A 3D Convolutional Neural Network for real-time object class recognition2016
- Volumetric and Multi-view CNNs for Object Classification on 3D Data2016
- Pairwise Decomposition of Image Sequences for Active Multi-view Recognition2016
- VoxNet: A 3D Convolutional Neural Network for real-time object recognition2015
- Fine-grained semi-supervised labeling of large shape collectionsACM Transactions on Graphics, 2013
- Hough Transform and 3D SURF for Robust Three Dimensional Classification2010
- On visual similarity based 2D drawing retrievalComputer-Aided Design, 2005
- Shape distributionsACM Transactions on Graphics, 2002