arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.12980cs.CV
收藏

一个新的任务人-物交互数据集与NVS基准

A New People-Object Interaction Dataset and NVS Benchmarks

Institute of Image Communication and Network Engineering, Shanghai Jiao Tong University(上海交通大学图像通信与网络工程研究所);School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University(上海交通大学电子信息与电气工程学院);SJTU-ParisTech Elite Institute of Technology, Shanghai Jiao Tong University(上海交通大学巴黎卓越工程师学院);Cooperative Medianet Innovation Center, Shanghai Jiao Tong University(上海交通大学合作媒体网络创新中心)

Shuai Guo, Houqiang Zhong, Qiuwen Wang, Ziyu Chen, Yijie Gao, Jiajing Yuan, Chenyu Zhang, Rong Xie, Li Song

AI总结:

本文提出一个包含38个系列30视角RGB-D视频序列的新人-物交互数据集,并建立NVS基准以促进相关研究。

摘要:

最近,人-物交互场景中的NVS受到了越来越多的关注。现有的人-物交互数据集主要由有限视角的静态数据组成,仅提供RGB图像或视频,大多包含单个人与物体的交互。此外,这些数据集在光照环境方面存在复杂性,同步性差,分辨率低,阻碍了高质量的人-物交互研究。在本文中,我们引入了一个新的人-物交互数据集,包含38个系列的30视角多人或单人RGB-D视频序列,并附带相机参数、前景掩码、SMPL模型、一些点云和网格文件。视频序列由30台Kinect Azures捕获,均匀环绕场景,每台以4K分辨率25 FPS录制,持续1到19秒。同时,我们在我们的数据集上评估了一些SOTA NVS模型,以建立NVS基准。我们希望我们的工作能激发人-物交互领域的进一步研究。

英文摘要:

Recently, NVS in human-object interaction scenes has received increasing attention. Existing human-object interaction datasets mainly consist of static data with limited views, offering only RGB images or videos, mostly containing interactions between a single person and objects. Moreover, these datasets exhibit complexities in lighting environments, poor synchronization, and low resolution, hindering high-quality human-object interaction studies. In this paper, we introduce a new people-object interaction dataset that comprises 38 series of 30-view multi-person or single-person RGB-D video sequences, accompanied by camera parameters, foreground masks, SMPL models, some point clouds, and mesh files. Video sequences are captured by 30 Kinect Azures, uniformly surrounding the scene, each in 4K resolution 25 FPS, and lasting for 1$\sim$19 seconds. Meanwhile, we evaluate some SOTA NVS models on our dataset to establish the NVS benchmarks. We hope our work can inspire further research in humanobject interaction.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑