arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2311.03312cs.CV
收藏

单个带上下文的2D姿态对3D人体姿态估计的价值抵得上数百个

A Single 2D Pose with Context is Worth Hundreds for 3D Human Pose Estimation

Robotics Institute, Carnegie Mellon University(卡内基梅隆大学机器人研究所);Center for Research in Computer Vision, University of Central Florida(中佛罗里达大学计算机视觉研究中心);Key Laboratory of Machine Perception, Peking University, Shenzhen Graduate School(北京大学深圳研究生院机器感知重点实验室)

Qitao Zhao, Ce Zheng, Mengyuan Liu, Chen Chen

AI总结:

针对3D人体姿态估计主流时序范式的性能饱和、计算量大等问题,提出利用预训练2D姿态检测器的中间视觉表示获取空间上下文的Context-Aware PoseFormer,单帧性能超越使用数百帧的SOTA方法。

摘要:

将2D姿态序列提升至3D的3D人体姿态估计主流范式,严重依赖长期时序线索(即使用数量极多的视频帧)来提升精度,这会导致性能饱和、计算棘手以及非因果问题。这可归因于它们固有的空间上下文感知能力不足,因为普通2D关节坐标不携带视觉线索。为解决该问题,我们提出了一种简单却强大的方案:利用现成(预训练)的2D姿态检测器生成的、可直接获取的中间视觉表示——甚至无需针对3D任务进行微调。关键观察结果是,当姿态检测器学习定位2D关节时,由于骨干网络中的区域运算,这类表示(例如特征图)会隐式编码以关节为中心的空间上下文。我们设计了一个名为Context-Aware PoseFormer的简单基线来展示其有效性。在不获取任何时序信息的情况下,所提方法在速度和精度上均显著优于无上下文的对应方法PoseFormer,以及使用多达数百帧视频的其他SOTA方法。项目页面:https://qitaozhao.github.io/ContextAware-PoseFormer

英文摘要:

The dominant paradigm in 3D human pose estimation that lifts a 2D pose sequence to 3D heavily relies on long-term temporal clues (i.e., using a daunting number of video frames) for improved accuracy, which incurs performance saturation, intractable computation and the non-causal problem. This can be attributed to their inherent inability to perceive spatial context as plain 2D joint coordinates carry no visual cues. To address this issue, we propose a straightforward yet powerful solution: leveraging the readily available intermediate visual representations produced by off-the-shelf (pre-trained) 2D pose detectors -- no finetuning on the 3D task is even needed. The key observation is that, while the pose detector learns to localize 2D joints, such representations (e.g., feature maps) implicitly encode the joint-centric spatial context thanks to the regional operations in backbone networks. We design a simple baseline named Context-Aware PoseFormer to showcase its effectiveness. Without access to any temporal information, the proposed method significantly outperforms its context-agnostic counterpart, PoseFormer, and other state-of-the-art methods using up to hundreds of video frames regarding both speed and precision. Project page: https://qitaozhao.github.io/ContextAware-PoseFormer

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。

↑