We propose a space-time Markov Random Field (MRF)
model to detect abnormal activities in video. The nodes in
the MRF graph correspond to a grid of local regions in the
video frames, and neighboring nodes in both space and time
are associated with links. To learn normal patterns of activ-
ity at each local node, we capture the distribution of its typ-
ical optical flow with a Mixture of Probabilistic Principal
Component Analyzers. For any new optical flow patterns
detected in incoming video clips, we use the learned model
and MRF graph to compute a maximum a posteriori esti-
mate of the degree of normality at each local node. Further,
we show how to incrementally update the current model pa-
rameters as new video observations stream in, so that the
model can efficiently adapt to visual context changes over
a long period of time. Experimental results on surveillance
videos show that our space-time MRF model robustly de-
tects abnormal activities both in a local and global...