The perturbations introduce various degrees of background and partiality to objects. In this work, we use four perturbation variants in the increasing order of difficulty: PB T25, PB T25 R, PB T50 R, and PB T50 RS. Suffix T25 and T50 denote translation that randomly shifts the bounding box up to 25% and 50% of its size from the box centroid along each world axis. Suffix R and S denotes rotation and scaling. Each perturbation variant contains five random samples for each original object, resulting in up to 14, 510 perturbed objects in total. Since perturbation mightintroduce invalid objects, e.g., objects that are almost completely out of the bounding box of interest, we perform an additional check after perturbation by ensuring that at least 50% of the original object points remain in the boundingbox. Objects that do not satisfy this condition are discarded. Sample point clouds of these variants are shown in Figure 3. More details about perturbing objects can be found in our supplementary material.
五种分类，越来越难。其中每一种又分了5种train/test split ，其中每一种又有 存在背景和去除背景的。所以一共有 5（5类 ） 5（五种split）2(有无背景) 50个数据集， 考虑的也是原始点云2048点的三维坐标，无引入其他信息。
split分类是因为去除因为split不同，而导致最终结果不同，多次实验取平均值。
5类点云，是来验证模型的鲁棒性，对于有裁切，或者与物体无关的一些under cover upper cover的影响。
2类，是验证背景点云对于点云分类来说是噪声。
有背景的数据，每一个点云还有一个mask数据，其中代表了一个点是属于背景点，还是物体点。
如表6、7、8和9所示，我们在ObjectNet数据集上包括了来自其他四个训练/测试拆分的结果，此外，如表10所示，我们还包括了针对背景点研究的这些其他拆分的结果，并且 再次表明，在我们最困难的变体中，背景点向网络学习引入了噪声，如w / BG案例在不同拆分中的较低性能所描绘的。
The results are 2902 objects that are categorized into 15 categories. The raw objects are represented by a list of points with global and local coordinates, normals, colors attributes and semantic labels，Other works synthesize challenges on CAD data by introducing noise simulated by Gaussians [4, 12] or created with a parametic model [6]. Recently, thetrend of sim2real [3] also aims to bridge the gap between synthetic and real data. As in the experiment with synthetic data, we sample all raw objects to 1024 points as input to the networks and all methods were trained using only the local (x, y, z) coordinates. We will make our dataset publicly available for future research. Table 2 summarizes classes and objects in our dataset. 结果是2902个对象，分为15类。 原始对象由具有全局和局部坐标，法线，颜色属性和语义标签的点列表表示。其他作品通过引入高斯模拟[4、12]或用参数模型创建的噪声[6]合成了CAD数据中的挑战。 最近，sim2real [3]的趋势也旨在弥合合成数据和真实数据之间的差距。 与使用合成数据进行的实验一样，我们将所有原始对象采样到1024点作为网络的输入，并且所有方法都仅使用局部（x，y，z）坐标进行了训练。 我们将公开我们的数据集以供将来研究。 表2总结了我们数据集中的类和对象。
Based on the selected objects, we construct several variants that represent different levels of difficulty of our dataset. This allows us to explore the robustness of existing classification methods in more extreme real-world scenarios. Vanilla. The first variant is referred to as OBJ ONLY which includes only ground truth segmented objects extracted from the scene meshes datasets. This variant has the closest form analogous to its CAD counterpart, and is used to investigate the robustness of classification methods to noisy objects with deformed geometric shape and nonuniform surface density. Sample objects of this variant are shown in Figure 2(a). Background. The previous variant assumes that an object can be accurately segmented before being classified. However, in real-world scans, objects are often presented in under-segmentation situations, i.e., background elements or parts of nearby objects are included, and accurate annotations for such under-segmentations are also not always available. Those background elements may provide the context where objects belong to, and thus would become a good hint for object classification, e.g., laptops often sit on desks. However, they may also introduce distractions which corrupt (a) Objects only. (b) Objects with background. Figure 2. Example objects from our dataset. the classification, e.g., a pen may be under-segmented with a table where it sits on and thus could be considered as a part of the table rather than a separate object. To study these factors, we introduce a variant of our dataset where objects are attached with background data (OBJ BG). We determine such background by using the ground truth axis-aligned object bounding boxes. Specifically, given a bounding box, all points in the box are extracted to form an object. Sample objects with background are shown in Figure 2(b). Perturbed. The given bounding boxes from the groundtruth tightly enclose the objects. However, in real-world scenarios bounding boxes may over- or under-cover, or even split objects. For example, in object detection techniques such as R-CNN [13], object category has to be predicted from a rough bounding box that localizes a candidate object. To simulate this challenge, we extend our dataset by translating, rotating (about the gravity axis), and scaling the ground truth bounding boxes before extracting the geometry in the box. We name the variants of these perturbations with a common prefix PB.
基于选定的对象，我们构造了几个变量，分别表示数据集的不同难度。这使我们能够在更极端的实际场景中探索现有分类方法的鲁棒性。香草。第一种变体仅称为OBJ，它仅包含从场景网格数据集中提取的地面真相分段对象。此变量的形式最接近于其CAD对应变量，用于研究分类方法对具有变形的几何形状和不均匀的表面密度的嘈杂对象的鲁棒性。图2（a）中显示了此变体的示例对象。背景。先前的变型假设可以在对对象进行分类之前对其进行准确的分割。但是，在现实世界的扫描中，通常在分割不足的情况下呈现对象，即，包括背景元素或附近物体的一部分，并且这种分割不足的精确注释也不总是可用。这些背景元素可以提供对象所属的上下文，因此将成为对象分类的良好提示，例如，笔记本电脑经常坐在桌子上。但是，它们也可能引起干扰，这些干扰只会破坏（a）对象。 （b）有背景的物体。图2.我们数据集中的示例对象。分类（例如笔）可能在其所在的桌子上被细分，因此可以被视为桌子的一部分，而不是单独的对象。为了研究这些因素，我们介绍了数据集的一种变体，其中对象与背景数据（OBJ BG）相连。我们通过使用地面真轴对齐的对象边界框来确定此类背景。具体来说，给定边界框，提取框内的所有点以形成对象。具有背景的示例对象如图2（b）所示。忐忑。来自地面真相的给定边界框将对象紧密包裹起来。但是，在实际情况下，边界框可能会被掩盖或掩盖，甚至被分割。例如，在诸如R-CNN [13]之类的对象检测技术中，必须从定位候选对象的粗糙边界框预测对象类别。为了模拟这一挑战，我们通过在平移，旋转（绕重力轴）和缩放地面真相边界框之前扩展数据集，然后再提取框中的几何图形。我们用公共前缀PB命名这些扰动的变体。