Multi-information-based convolutional neural network with attention mechanism for pedestrian trajectory prediction

作者：

Highlights：

• We propose to combine the pose and the 2D-3D size information of the pedestrian to indicate human intention.

• We propose to use depth map to extract human-human and human-scene interactions for future trajectory prediction.

• Compared with LSTM, our model speeds up the calculation process through the parallel computing feature of convolution.

摘要

•We propose to combine the pose and the 2D-3D size information of the pedestrian to indicate human intention.•We propose to use depth map to extract human-human and human-scene interactions for future trajectory prediction.•Compared with LSTM, our model speeds up the calculation process through the parallel computing feature of convolution.

论文关键词：Depth map,Pose,2D-3D size information,Convolutional neural network,Trajectory prediction

论文评审过程：Received 18 December 2020, Revised 4 January 2021, Accepted 11 January 2021, Available online 23 January 2021, Version of Record 2 February 2021.

论文官网地址：https://doi.org/10.1016/j.imavis.2021.104110