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

TL;DR

This paper advances 3D object classification by developing improved volumetric and multi-view CNN architectures, achieving state-of-the-art results through extensive analysis and novel multi-resolution filtering techniques.

Contribution

Introduces two new volumetric CNN architectures and a multi-resolution filtering approach for multi-view CNNs, enhancing 3D object classification performance.

Findings

Outperforms existing state-of-the-art methods

Provides extensive analysis of design choices

Demonstrates the effectiveness of multi-resolution filtering

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-the-art 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 multi-resolution 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.