Adversarially Learned Abnormal Trajectory Classifier
This addresses the problem of automating abnormal event detection in surveillance and traffic monitoring without manual thresholds, though it is incremental as it adapts existing GAN frameworks.
The paper tackles abnormal event detection from trajectory data by proposing an adversarial deep neural network classifier trained unsupervised, achieving the best accuracy for abnormal trajectory detection in urban traffic videos and demonstrating generalization on the CAVIAR dataset.
We address the problem of abnormal event detection from trajectory data. In this paper, a new adversarial approach is proposed for building a deep neural network binary classifier, trained in an unsupervised fashion, that can distinguish normal from abnormal trajectory-based events without the need for setting manual detection threshold. Inspired by the generative adversarial network (GAN) framework, our GAN version is a discriminative one in which the discriminator is trained to distinguish normal and abnormal trajectory reconstruction errors given by a deep autoencoder. With urban traffic videos and their associated trajectories, our proposed method gives the best accuracy for abnormal trajectory detection. In addition, our model can easily be generalized for abnormal trajectory-based event detection and can still yield the best behavioural detection results as demonstrated on the CAVIAR dataset.