People Counting in Videos by Fusing Temporal Cues from Spatial Context-Aware Convolutional Neural Networks | Publicación