RepNet: Weakly Supervised Training of an Adversarial Reprojection Network for 3D Human Pose Estimation
RepNet: Weakly Supervised Training of an Adversarial Reprojection Network for 3D Human Pose Estimation
复制标题
DOI：
10.1109/cvpr.2019.00797
复制
发表时间：
2019-02
期刊：
影响因子：
--
通讯作者：
Bastian Wandt;B. Rosenhahn
中科院分区：
文献类型：
--
作者：
Bastian Wandt;B. Rosenhahn
本文解决了从单张图像进行 3D 人体姿势估计的问题。虽然很长一段时间以来，人体骨骼都是通过满足重投影误差来参数化并适应观察结果的，但现在研究人员直接使用神经网络从观察结果中推断出 3D 姿势。然而，这些方法中的大多数都忽略了必须满足重投影约束并且对过度拟合敏感的事实。我们通过忽略 2D 到 3D 对应关系来解决过度拟合问题。这有效地避免了训练数据的简单记忆，并允许弱监督训练。所提出的重投影网络 (RepNet) 的一部分使用对抗性训练方法学习从 2D 姿势分布到 3D 姿势分布的映射。网络的另一部分估计相机。这允许定义一个网络层，该网络层将估计的 3D 姿势重投影回 2D，从而产生重投影损失函数。我们的实验表明，RepNet 对未知数据具有很好的泛化能力，并且在应用于未知数据时优于最先进的方法。此外，我们的实现在标准台式电脑上实时运行。
This paper addresses the problem of 3D human pose estimation from single images. While for a long time human skeletons were parameterized and fitted to the observation by satisfying a reprojection error, nowadays researchers directly use neural networks to infer the 3D pose from the observations. However, most of these approaches ignore the fact that a reprojection constraint has to be satisfied and are sensitive to overfitting. We tackle the overfitting problem by ignoring 2D to 3D correspondences. This efficiently avoids a simple memorization of the training data and allows for a weakly supervised training. One part of the proposed reprojection network (RepNet) learns a mapping from a distribution of 2D poses to a distribution of 3D poses using an adversarial training approach. Another part of the network estimates the camera. This allows for the definition of a network layer that performs the reprojection of the estimated 3D pose back to 2D which results in a reprojection loss function. Our experiments show that RepNet generalizes well to unknown data and outperforms state-of-the-art methods when applied to unseen data. Moreover, our implementation runs in real-time on a standard desktop PC.