Abstract
Numerous prior studies predominantly emphasize constructing relation vectors for individual neighborhood points and generating dynamic kernels for each vector and embedding these into high-dimensional spaces to capture implicit local structures. However, we contend that such implicit high-dimensional structure modeling approch inadequately represents the local geometric structure of point clouds due to the absence of explicit structural information. Hence, we introduce X-3D, an explicit 3D structure modeling approach. X-3D functions by capturing the explicit local structural information within the input 3D space and employing it to produce dynamic kernels with shared weights for all neighborhood points within the current local region. This modeling approach introduces effective geometric prior and significantly diminishes the disparity between the local structure of the embedding space and the original input point cloud, thereby improving the extraction of local features. Experiments show that our method can be used on a variety of methods and achieves state-of-the-art performance on segmentation, classification, detection tasks with lower extra computational cost, such as \textbf{90.7\%} on ScanObjectNN for classification, \textbf{79.2\%} on S3DIS 6 fold and \textbf{74.3\%} on S3DIS Area 5 for segmentation, \textbf{76.3\%} on ScanNetV2 for segmentation and \textbf{64.5\%} mAP , \textbf{46.9\%} mAP on SUN RGB-D and \textbf{69.0\%} mAP , \textbf{51.1\%} mAP on ScanNetV2 . Our code is available at \href{this https URL}{this https URL}.
Abstract (translated)
许多先前的研究主要侧重于为单个聚类点构建关系向量并生成动态核，并将它们嵌入高维空间以捕捉隐含的局部结构。然而，我们认为，由于缺乏明确的结构信息，这种隐含的高维结构建模方法不足以代表点云的局部几何结构。因此，我们引入了X-3D，一种明确的3D结构建模方法。X-3D通过捕获输入3D空间中的显式局部结构信息并使用它来生成共享权重的动态核来工作。这种建模方法引入了有效的几何先验，并显著降低了嵌入空间中局部结构与原始输入点云之间的差异，从而提高了局部特征的提取。实验表明，我们的方法可以应用于各种方法，并且在分类、检测任务上具有与较低附加计算成本 state-of-the-art 性能，例如在 ScanObjectNN 上达到 90.7% 的分类精度，在 S3DIS 6 折和 S3DIS Area 5 上达到 79.2% 的检测精度，在 ScanNetV2 上达到 74.3% 的检测精度和在 ScanNetV2 上达到 64.5% 的mAP，在 SUN RGB-D 上达到 46.9% 的mAP，在 ScanNetV2 上达到 69.0% 的mAP，在 ScanNetV2 上达到 51.1% 的mAP。我们的代码可在此处下载：https://this https URL/this https URL。
URL
https://arxiv.org/abs/2404.15010
