arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2311.04591cs.CVcs.MMcs.ROeess.IV
收藏

探索基于事件的3D事件表示的人体姿态估计

Exploring Event-based Human Pose Estimation with 3D Event Representations

Zhejiang University(浙江大学);Shanghai SUPREMIND Technology Company Ltd.(上海思普锐明科技有限公司);Hunan University(湖南大学)

Xiaoting Yin, Hao Shi, Jiaan Chen, Ze Wang, Yaozu Ye, Kailun Yang, Kaiwei Wang

AI总结:

针对传统相机在极端场景下可靠性下降及现有方法忽略事件异步性的问题,提出两种3D事件表示(RasEPC和DEV)以降低计算需求并提取3D线索,同时发布合成数据集EV-3DPW,在多个数据集上验证了方法的有效性。

摘要:

人体姿态估计是计算机视觉中一项基础且吸引人的任务。尽管传统相机被广泛应用,但它们在高动态范围或严重运动模糊场景下的可靠性会下降,而事件相机提供了一种稳健的解决方案。主流的基于事件的方法将事件累积成帧,忽略了对于区分不同动作至关重要的异步性和高时间分辨率。为了解决这一问题并释放事件信息的3D潜力,我们引入了两种3D事件表示:光栅化事件点云(RasEPC)和解耦事件体素(DEV)。RasEPC在相同位置上聚合简洁时间切片内的事件,保留了它们的3D属性以及统计信息,从而显著降低了内存和计算需求。同时,DEV表示将事件离散化为体素,并将其投影到三个正交平面上,利用解耦事件注意力从2D平面中提取3D线索。此外,我们开发并发布了EV-3DPW,这是一个合成的基于事件的数据集,旨在促进户外场景的训练和定量分析。我们的方法在DHP19公开数据集、MMHPSD数据集和我们的EV-3DPW数据集上进行了测试,并通过衍生的驾驶场景数据集EV-JAAD和户外采集车辆进行了进一步的定性验证。我们的代码和数据集已在 https://github.com/MasterHow/EventPointPose 公开提供。

英文摘要:

Human pose estimation is a fundamental and appealing task in computer vision. Although traditional cameras are commonly applied, their reliability decreases in scenarios under high dynamic range or heavy motion blur, where event cameras offer a robust solution. Predominant event-based methods accumulate events into frames, ignoring the asynchronous and high temporal resolution that is crucial for distinguishing distinct actions. To address this issue and to unlock the 3D potential of event information, we introduce two 3D event representations: the Rasterized Event Point Cloud (RasEPC) and the Decoupled Event Voxel (DEV). The RasEPC aggregates events within concise temporal slices at identical positions, preserving their 3D attributes along with statistical information, thereby significantly reducing memory and computational demands. Meanwhile, the DEV representation discretizes events into voxels and projects them across three orthogonal planes, utilizing decoupled event attention to retrieve 3D cues from the 2D planes. Furthermore, we develop and release EV-3DPW, a synthetic event-based dataset crafted to facilitate training and quantitative analysis in outdoor scenes. Our methods are tested on the DHP19 public dataset, MMHPSD dataset, and our EV-3DPW dataset, with further qualitative validation via a derived driving scene dataset EV-JAAD and an outdoor collection vehicle. Our code and dataset have been made publicly available at https://github.com/MasterHow/EventPointPose.

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。

↑