We are witnessing a proliferation of textured 3D models captured from the real world with automatic photo-reconstruction tools by people and professionals without a proper technical background in computer graphics. Digital 3D models of this class come with a unique set of characteristics and defects – especially concerning their parametrization – setting them starkly apart from 3D models originating from other, more traditional, sources. We study this class of 3D models by collecting a significant number of representatives and quantitatively evaluating their quality according to several metrics. These include a new invariant metric we carefully design to assess the amount of fragmentation of the UV map, which is one of the main weaknesses potentially hindering the usability of these models. Our results back the widely shared notion that models of this new class are still not fit for direct use in downstream applications (such as videogames), and require challenging processing steps. Regrettably, existing automatic geometry processing tools are not always up to the task: for example, we verify that the available tools for UV optimization often fail due to mesh inconsistencies, geometric and topological noise, excessive resolution, or other factors; moreover, even when an output is produced, it rarely represents a significant improvement over the input (according to the aforementioned measures). Therefore, we argue that further advancements are required by the computer graphics and geometry processing communities specifically targeted at this class of models. Towards this goal, we share the models we collected in this study as a new public repository, Real-World Textured Things (RWTT), intended as a benchmark to systematic field-test and compare future algorithms. RWTT consists of 568 carefully selected textured 3D models representative of the most popular photo-reconstruction tools currently available. We also provide a web interface to browse the dataset by the metadata we collected during our experiments and a tool, TexMetro, to compute the same set of measures on generic UV mapped datasets.

我们目睹了在没有适当计算机图形技术背景的情况下，人们和专业人员使用自动照片重建工具从现实世界中捕获的纹理3D模型的激增。此类数字3D模型具有一组独特的特征和缺陷（尤其是在参数化方面），使其与源自其他传统资源的3D模型形成鲜明的区别。我们通过收集大量代表并根据几种指标定量评估其质量来研究此类3D模型。这些包括一个新的不变式度量标准，我们精心设计以评估UV贴图的碎片数量，这是可能阻碍这些模型使用的主要弱点之一。我们的结果支持了广泛认同的观念，即这种新类型的模型仍不适合直接用于下游应用程序（例如视频游戏），并且需要艰巨的处理步骤。遗憾的是，现有的自动几何处理工具并不总是能够胜任该任务：例如，我们验证了用于UV优化的可用工具通常会由于网格不一致，几何和拓扑噪声，过大的分辨率或其他因素而失败；此外，即使产生了产出，也很少代表对投入的重大改进（根据上述措施）。因此，我们认为，专门针对此类模型的计算机图形和几何处理社区需要进一步的发展。为了实现这一目标，我们将在这项研究中收集的模型共享为一个新的公共资源库，真实纹理物体（RWTT），旨在作为系统性现场测试和比较未来算法的基准。RWTT由568种精心挑选的纹理3D模型组成，这些模型代表了当前最流行的照片重建工具。我们还提供了一个Web界面，可通过我们在实验中收集的元数据浏览数据集，并提供一个工具TexMetro，以对通用UV映射数据集计算相同的一组度量。