Volumetric and Multi-view CNNs for Object Classification on 3D Data

Citations Over TimeTop 1% of 2016 papers

Abstract

3D shape models are becoming widely available and easier to capture, making available 3D information crucial for progress in object classification. Current state-of-theart methods rely on CNNs to address this problem. Recently, we witness two types of CNNs being developed: CNNs based upon volumetric representations versus CNNs based upon multi-view representations. Empirical results from these two types of CNNs exhibit a large gap, indicating that existing volumetric CNN architectures and approaches are unable to fully exploit the power of 3D representations. In this paper, we aim to improve both volumetric CNNs and multi-view CNNs according to extensive analysis of existing approaches. To this end, we introduce two distinct network architectures of volumetric CNNs. In addition, we examine multi-view CNNs, where we introduce multiresolution filtering in 3D. Overall, we are able to outperform current state-of-the-art methods for both volumetric CNNs and multi-view CNNs. We provide extensive experiments designed to evaluate underlying design choices, thus providing a better understanding of the space of methods available for object classification on 3D data.

Related Papers

• → Deep Residual Learning for Image Recognition(2016)216,943 cited
• → VoxNet: A 3D Convolutional Neural Network for real-time object recognition(2015)3,536 cited
• → OctNet: Learning Deep 3D Representations at High Resolutions(2017)1,669 cited
• → PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric\n Space(2017)7,051 cited
• → PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation(2016)2,879 cited