Several applications such as autonomous driving, augmented reality and virtual reality requires a precise prediction of the 3D human pose. Recently, a new problem was introduced in the field to predict the 3D human poses from an observed 2D poses. We propose Skeleton-Graph, a deep spatio-temporal graph CNN model that predicts the future 3D skeleton poses in a single pass from the 2D ones. Unlike prior works, Skeleton-Graph focuses on modeling the interaction between the skeleton joints by exploiting their spatial configuration. This is being achieved by formulating the problem as a graph structure while learning a suitable graph adjacency kernel. By the design, Skeleton-Graph predicts the future 3D poses without divergence on the long-term unlike prior works. We also introduce a new metric that measures the divergence of predictions on the long-term. Our results show an FDE improvement of at least 27% and an ADE of 4% on both the GTA-IM and PROX datasets respectively in comparison with prior works. Also, we are 88% and 93% less divergence on the long-term motion prediction in comparison with prior works on both GTA-IM and PROX datasets. https://github.com/abduallahmohamed/Skeleton-Graph.git
翻译:自主驱动、 增强现实和虚拟现实等几个应用程序需要精确预测 3D 人形。 最近, 实地出现了一个新问题, 以预测观察到的 2D 外表的 3D 人形。 我们提出Skeleton- Graph, 这是一种远方平面图 CNN 模型, 预测未来 3D 骨架在与 2D 相隔的单一通道中形成。 不同于先前的工程, Skeleton- graph 侧重于通过利用其空间配置来模拟骨干关节之间的相互作用。 这是通过将问题作为图表结构来表述,同时学习适当的图形相邻内核。 根据设计, Skeleton- Graph 预测未来 3D 与先前的工程不同, 未来 3D 与以往的工程没有差异。 我们还引入了一个新的衡量长期预测差异的尺度。 我们的结果表明, GTA-IM- IM / PROAmbhed 数据集的改善率至少为 27%, 与 4% 。 另外, 与之前的工作相比, 在长期 IM/ PROABS/ GRAMS) 。