Point clouds are now commonly used to represent 3D scenes in virtual world, in addition to 3D meshes. Their ease of capture enable various applications on mobile devices, such as smartphones or other microcontrollers. Point cloud compression is now at an advanced level and being standardized. Nevertheless, quality assessment databases, which is needed to develop better objective quality metrics, are still limited. In this work, we create a broad quality assessment database for static point clouds, mainly for telepresence scenario. For the sake of completeness, the created database is analyzed using the mean opinion scores, and it is used to benchmark several state-of-the-art quality estimators. The generated database is named Broad quality Assessment of Static point clouds In Compression Scenario (BASICS). Currently, the BASICS database is used as part of the ICIP 2023 Grand Challenge on Point Cloud Quality Assessment, and therefore only a part of the database has been made publicly available at the challenge website. The rest of the database will be made available once the challenge is over.
翻译:点云现在已被广泛用于虚拟世界中表示3D场景,与3D网格相辅相成。其易于采集的特性使得在智能手机或其他微控制器等移动设备上实现多种应用成为可能。点云压缩技术现已达到先进水平并正在标准化进程中。然而,用于开发更优客观质量指标所需的质量评估数据库仍十分有限。在本研究中,我们构建了一个针对静态点云(主要面向远程呈现场景)的广度质量评估数据库。为追求完备性,该数据库通过平均意见得分进行分析,并用于评估多种最新质量估算方法。生成的数据库命名为"压缩场景下静态点云的广度质量评估"(BASICS)。目前BASICS数据库已作为ICIP 2023点云质量评估大挑战的组成部分,因此在竞赛网站上仅公开了部分数据。剩余部分将在竞赛结束后公布。