Interpretation of state sequences in HMM for activity representation

We propose a method for activity representation based on seman-tic events, using the HMM framework. For every time instant, the
probability of event occurrence is computed by exploring a subset
of state sequences. The idea is that while activity trajectories may
have large variations at the data or the state levels, they may ex-
hibit similarities at the event level. Our experiments show the ap-
plication of these events to activity recognition in an office envi-
ronment and to anomalous trajectory detection using surveillance
video data.