Based on human bone joints, skeleton information has clear and simple features is not easily affected by appearance factors. In this paper, an improved feature of Gist, ExGist, proposed to describe the joints for action recognition. The joint coordinates are extracted using OpenPose thermodynamic diagram, ExGist used extraction. advantage that it can effectively characterize local global while ...

Graph convolutional networks (GCNs) are widely adopted in skeleton-based action recognition due to their powerful ability model data topology. We argue that the performance of recent proposed methods is limited by following factors. First, predefined graph structures shared throughout network, lacking flexibility and capacity multi-grain semantic information. Second, relations among global join...

For action recognition learning, 2D CNN-based methods are efficient but may yield redundant features due to applying the same convolution kernel each frame. Recent efforts attempt capture motion information by establishing inter-frame connections while still suffering limited temporal receptive field or high latency. Moreover, feature enhancement is often only performed channel space dimension ...