Sequential keypoint density estimator: an overlooked baseline of skeleton-based video anomaly detection
This work addresses anomaly detection in safety-critical applications like surveillance and healthcare, though it is incremental as it builds on existing skeleton-based methods.
The authors tackled video anomaly detection by modeling human skeleton sequences with an autoregressive keypoint density estimator, achieving state-of-the-art results on UBnormal and MSAD-HR datasets and competitive performance on ShanghaiTech.
Detecting anomalous human behaviour is an important visual task in safety-critical applications such as healthcare monitoring, workplace safety, or public surveillance. In these contexts, abnormalities are often reflected with unusual human poses. Thus, we propose SeeKer, a method for detecting anomalies in sequences of human skeletons. Our method formulates the skeleton sequence density through autoregressive factorization at the keypoint level. The corresponding conditional distributions represent probable keypoint locations given prior skeletal motion. We formulate the joint distribution of the considered skeleton as causal prediction of conditional Gaussians across its constituent keypoints. A skeleton is flagged as anomalous if its keypoint locations surprise our model (i.e. receive a low density). In practice, our anomaly score is a weighted sum of per-keypoint log-conditionals, where the weights account for the confidence of the underlying keypoint detector. Despite its conceptual simplicity, SeeKer surpasses all previous methods on the UBnormal and MSAD-HR datasets while delivering competitive performance on the ShanghaiTech dataset.