Crowd Event Perception Based on Spatiotemporal Weber Field
Figure 9
The qualitative results of the crowd behavior perception for four sample videos. (a) are sample results from Seq. 1, and (b) are sample results from Seq. 2. The bars are utilized to represent the labels of each frame for the videos in the sample video, and different colors represent the different corresponding crowd behaviors. The bars indicate the ground truth, the detection results based on the proposed spatiotemporal Weber field, social force model [24], HOG/HOF [36], HOG3D [37], and pure optical flow [8], respectively.
