Normal Learning in Videos with Attention Prototype Network
This work addresses the problem of memory inefficiency and lack of adaptability in video anomaly detection for surveillance and security applications, offering an incremental improvement over existing methods.
The paper tackles video anomaly detection by proposing a self-attention prototype unit (APU) to encode normal patterns without extra memory cost and a circulative attention unit (CAU) for fast adaptation to new scenes, achieving superior performance over state-of-the-art methods on various benchmarks.
Frame reconstruction (current or future frame) based on Auto-Encoder (AE) is a popular method for video anomaly detection. With models trained on the normal data, the reconstruction errors of anomalous scenes are usually much larger than those of normal ones. Previous methods introduced the memory bank into AE, for encoding diverse normal patterns across the training videos. However, they are memory consuming and cannot cope with unseen new scenarios in the testing data. In this work, we propose a self-attention prototype unit (APU) to encode the normal latent space as prototypes in real time, free from extra memory cost. In addition, we introduce circulative attention mechanism to our backbone to form a novel feature extracting learner, namely Circulative Attention Unit (CAU). It enables the fast adaption capability on new scenes by only consuming a few iterations of update. Extensive experiments are conducted on various benchmarks. The superior performance over the state-of-the-art demonstrates the effectiveness of our method. Our code is available at https://github.com/huchao-AI/APN/.