Revisiting Point Cloud Shape Classification with a Simple and Effective Baseline
This work addresses performance evaluation issues in point cloud shape classification, showing that simple methods can be highly effective, which is incremental but impactful for the field.
The study found that auxiliary factors like evaluation schemes and data augmentation significantly affect point cloud classification performance, and a simple projection-based method (SimpleView) achieves on par or better results than state-of-the-art methods on ModelNet40 and outperforms them on ScanObjectNN.
Processing point cloud data is an important component of many real-world systems. As such, a wide variety of point-based approaches have been proposed, reporting steady benchmark improvements over time. We study the key ingredients of this progress and uncover two critical results. First, we find that auxiliary factors like different evaluation schemes, data augmentation strategies, and loss functions, which are independent of the model architecture, make a large difference in performance. The differences are large enough that they obscure the effect of architecture. When these factors are controlled for, PointNet++, a relatively older network, performs competitively with recent methods. Second, a very simple projection-based method, which we refer to as SimpleView, performs surprisingly well. It achieves on par or better results than sophisticated state-of-the-art methods on ModelNet40 while being half the size of PointNet++. It also outperforms state-of-the-art methods on ScanObjectNN, a real-world point cloud benchmark, and demonstrates better cross-dataset generalization. Code is available at https://github.com/princeton-vl/SimpleView.