Pedestrian trajectory prediction is an extremely challenging problem because of the crowdedness and clutter of the scenes. Previous deep learning LSTM-based approaches focus on the neighbourhood influence of pedestrians but ignore the scene layouts in pedestrian trajectory prediction. In this paper, a novel hierarchical LSTM-based network is proposed to consider both the influence of social neighbourhood and scene layouts. Our SS-LSTM, which stands for Social-Scene-LSTM, uses three different LSTMs to capture person, social and scene scale information. We also use a circular shape neighbourhood setting instead of the traditional rectangular shape neighbourhood in the social scale. We evaluate our proposed method against two baseline methods and a state-of-art technique on three public datasets. The results show that our method outperforms other methods and that using circular shape neighbourhood improves the prediction accuracy.

Video-based human action recognition is currently one of the most active research areas in computer vision. Various research studies indicate that the performance of action recognition is highly dependent on the type of features being extracted and how the actions are represented. Since the release of the Kinect camera, a large number of Kinect-based human action recognition techniques have been proposed in the literature. However, there still does not exist a thorough comparison of these Kinect-based techniques under the grouping of feature types, such as handcrafted versus deep learning features and depth-based versus skeleton-based features. In this paper, we analyze and compare ten recent Kinect-based algorithms for both cross-subject action recognition and cross-view action recognition using six benchmark datasets. In addition, we have implemented and improved some of these techniques and included their variants in the comparison. Our experiments show that the majority of methods perform better on cross-subject action recognition than cross-view action recognition, that skeleton-based features are more robust for cross-view recognition than depth-based features, and that deep learning features are suitable for large datasets.

Existing techniques for 3D action recognition are sensitive to viewpoint variations because they extract features from depth images which are viewpoint dependent. In contrast, we directly process point clouds for cross-view action recognition from unknown and unseen views. We propose the histogram of oriented principal components (HOPC) descriptor that is robust to noise, viewpoint, scale and action speed variations. At a 3D point, HOPC is computed by projecting the three scaled eigenvectors of the pointcloud within its local spatio-temporal support volume onto the vertices of a regular dodecahedron. HOPC is also used for the detection of spatiotemporal keypoints (STK) in 3D pointcloud sequences so that view-invariant STK descriptors (or Local HOPC descriptors) at these key locations only are used for action recognition. We also propose a global descriptor computed from the normalized spatio-temporal distribution of STKs in 4-D, which we refer to as STK-D. We have evaluated the performance of our proposed descriptors against nine existing techniques on two cross-view and three single-view human action recognition datasets. The experimental results show that our techniques provide significant improvement over state-of-the-art methods.