Temporally Consistent 3D Human Pose Estimation Using Dual 360° Cameras
Temporally Consistent 3D Human Pose Estimation Using Dual 360° Cameras
复制标题
DOI：
10.1109/wacv48630.2021.00013
复制
发表时间：
2020-11
期刊：
影响因子：
--
通讯作者：
Matthew Shere;Hansung Kim;A. Hilton
中科院分区：
文献类型：
--
作者：
Matthew Shere;Hansung Kim;A. Hilton
本文提出了一种 3D 人体姿势估计系统，该系统使用一对立体 360° 传感器从单个位置捕获完整场景。该方法结合了全向捕获、多视图 3D 位姿估计的准确性和单目采集的便携性的优点。关节位置的关节单目置信图是根据 360° 图像估计的，并用于将 3D 骨架拟合到每个帧。执行时间数据关联和平滑以在整个序列中生成准确的 3D 姿态估计。我们在 Panoptic Studio 数据集以及用于跟踪多人的真实 360° 视频上评估我们的系统，证明使用 30 厘米基线相机的平均每个关节位置误差为 12.47 厘米。当呈现有限的拍摄对象视图时，我们还展示了相对于透视和 360° 多视图系统的改进功能。
This paper presents a 3D human pose estimation system that uses a stereo pair of 360° sensors to capture the complete scene from a single location. The approach combines the advantages of omnidirectional capture, the accuracy of multiple view 3D pose estimation and the portability of monocular acquisition. Joint monocular belief maps for joint locations are estimated from 360° images and are used to fit a 3D skeleton to each frame. Temporal data association and smoothing is performed to produce accurate 3D pose estimates throughout the sequence. We evaluate our system on the Panoptic Studio dataset, as well as real 360° video for tracking multiple people, demonstrating an average Mean Per Joint Position Error of 12.47cm with 30cm baseline cameras. We also demonstrate improved capabilities over perspective and 360° multi-view systems when presented with limited camera views of the subject.