Abstract
There is considerable interest in techniques capable of iden- tifying anomalies and unusual events in busy outdoor scenes, e.g. road junctions. Many approaches achieve this by exploiting deviations in spa- tial appearance from some expected norm accumulated by a model over time. In this work we show that much can be gained from explicitly mod- elling temporal aspects in detail. Specifically, many traffic junctions are regulated by lights controlled by a timing device of considerable preci- sion, and it is in these situations that we advocate a model which learns periodic spatio-temporal patterns with a view to highlighting anomalous events such as broken-down vehicles, traffic accidents, or pedestrians jay- walking. More specifically, by estimating autocovariance of self-similarity, used previously in the context gait recognition, we characterize a scene by identifying a global fundamental period. As our model, we introduce a spatio-temporal grid of histograms built in accordance with some chosen feature. This model is then used to classify ob jects found in subsequent test data. In particular we demonstrate the effect of such characteriza- tion experimentally by monitoring the bounding box aspect ratio and optical flow field of ob jects detected on a road traffic junction, enabling our model to discriminate between people and cars sufficiently well to provide useful warnings of adverse behaviour in real time.