# Video Anomaly Detection by Estimating Likelihood of Representations

arXiv: https://arxiv.org/abs/2012.01468 · HTML: https://mlsanity.com/paper/2012.01468/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Dec 2020.

## Built on by (most cited first, 9 of 9)

- [Multi-Contextual Predictions with Vision Transformer for Video Anomaly Detection](https://mlsanity.com/paper/2206.08568/) (2022)
- [Video Anomaly Detection via Prediction Network with Enhanced Spatio-Temporal Memory Exchange](https://mlsanity.com/paper/2206.12914/) (2022)
- [Deep Crowd Anomaly Detection: State-of-the-Art, Challenges, and Future Research Directions](https://mlsanity.com/paper/2210.13927/) (2022)
- [Margin-Aware Intra-Class Novelty Identification for Medical Images](https://mlsanity.com/paper/2108.00117/) (2021)
- [A Survey on Video Analytics in Cloud-Edge-Terminal Collaborative Systems](https://mlsanity.com/paper/2502.06581/) (2025)
- [Look at Adjacent Frames: Video Anomaly Detection without Offline Training](https://mlsanity.com/paper/2207.13798/) (2022)
- [Bounding Boxes and Probabilistic Graphical Models: Video Anomaly Detection Simplified](https://mlsanity.com/paper/2407.06000/) (2024)
- [Video Anomaly Detection with Contours -- A Study](https://mlsanity.com/paper/2503.19588/) (2025)
- [Advancing Video Anomaly Detection: A Bi-Directional Hybrid Framework for Enhanced Single- and Multi-Task Approaches](https://mlsanity.com/paper/2504.14753/) (2025)

## Abstract (arXiv)

Video anomaly detection is a challenging task not only because it involves solving many sub-tasks such as motion representation, object localization and action recognition, but also because it is commonly considered as an unsupervised learning problem that involves detecting outliers. Traditionally, solutions to this task have focused on the mapping between video frames and their low-dimensional features, while ignoring the spatial connections of those features. Recent solutions focus on analyzing these spatial connections by using hard clustering techniques, such as K-Means, or applying neural networks to map latent features to a general understanding, such as action attributes. In order to solve video anomaly in the latent feature space, we propose a deep probabilistic model to transfer this task into a density estimation problem where latent manifolds are generated by a deep denoising autoencoder and clustered by expectation maximization. Evaluations on several benchmarks datasets show the strengths of our model, achieving outstanding performance on challenging datasets.