论文标题
3D分类的本地社区功能
Local Neighborhood Features for 3D Classification
论文作者
论文摘要
随着深度学习模型培训策略的进步，点云分类方法的训练正在显着改善。例如，PointNext在PointNet ++中采用了突出的培训技术和Invresnet层，在现实世界中的Scanobjectnn数据集上实现了超过7％的提高。但是，这些模型中的大多数都使用映射到更高维空间的邻域点的点坐标特征，同时忽略了在馈送到网络层之前计算的邻域特征。在本文中，我们重新访问了PointNext模型，以研究此类邻里点功能的使用和好处。我们在ModelNet40（合成），Scanobjectnn（现实世界）和最近的大型，现实世界中的杂货数据集（即3Dgrocery100）上训练和评估PointNext。此外，我们还提供了平均PointNext的顶部两个检查点的重量的额外推理策略，以提高分类精度。与上述想法一起，我们在带有现实世界数据集的PointNext模型上获得了0.5％，1％，4.8％，3.4％和1.6％的总体准确性，ScanObjectnn（最难的变体），3DGROCERY100'S APPER10，FROUITS，蔬菜，蔬菜和包装子集。我们还可以在ModelNet40上获得可比的0.2％精度增益。
With advances in deep learning model training strategies, the training of Point cloud classification methods is significantly improving. For example, PointNeXt, which adopts prominent training techniques and InvResNet layers into PointNet++, achieves over 7% improvement on the real-world ScanObjectNN dataset. However, most of these models use point coordinates features of neighborhood points mapped to higher dimensional space while ignoring the neighborhood point features computed before feeding to the network layers. In this paper, we revisit the PointNeXt model to study the usage and benefit of such neighborhood point features. We train and evaluate PointNeXt on ModelNet40 (synthetic), ScanObjectNN (real-world), and a recent large-scale, real-world grocery dataset, i.e., 3DGrocery100. In addition, we provide an additional inference strategy of weight averaging the top two checkpoints of PointNeXt to improve classification accuracy. Together with the abovementioned ideas, we gain 0.5%, 1%, 4.8%, 3.4%, and 1.6% overall accuracy on the PointNeXt model with real-world datasets, ScanObjectNN (hardest variant), 3DGrocery100's Apple10, Fruits, Vegetables, and Packages subsets, respectively. We also achieve a comparable 0.2% accuracy gain on ModelNet40.
