# MULDE: Multiscale Log-Density Estimation via Denoising Score Matching for Video Anomaly Detection

arXiv: https://arxiv.org/abs/2403.14497 · HTML: https://mlsanity.com/paper/2403.14497/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- Builds on 32 of its 51 references that are in the map.
- First posted on arXiv 21 Mar 2024.

## Builds on (most cited first, 30 of 32)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Generative Modeling by Estimating Gradients of the Data Distribution](https://mlsanity.com/paper/1907.05600/) (2019)
- [The Kinetics Human Action Video Dataset](https://mlsanity.com/paper/1705.06950/) (2017)
- [FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks](https://mlsanity.com/paper/1612.01925/) (2016)
- [Real-world Anomaly Detection in Surveillance Videos](https://mlsanity.com/paper/1801.04264/) (2018)
- [Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detection](https://mlsanity.com/paper/1904.02639/) (2019)
- [Future Frame Prediction for Anomaly Detection -- A New Baseline](https://mlsanity.com/paper/1712.09867/) (2017)
- [RMPE: Regional Multi-person Pose Estimation](https://mlsanity.com/paper/1612.00137/) (2016)
- [Hiera: A Hierarchical Vision Transformer without the Bells-and-Whistles](https://mlsanity.com/paper/2306.00989/) (2023)
- [Learning Memory-guided Normality for Anomaly Detection](https://mlsanity.com/paper/2003.13228/) (2020)
- [Learning Temporal Regularity in Video Sequences](https://mlsanity.com/paper/1604.04574/) (2016)
- [Why Normalizing Flows Fail to Detect Out-of-Distribution Data](https://mlsanity.com/paper/2006.08545/) (2020)
- [Anomaly Detection in Video via Self-Supervised and Multi-Task Learning](https://mlsanity.com/paper/2011.07491/) (2020)
- [A Hybrid Video Anomaly Detection Framework via Memory-Augmented Flow Reconstruction and Flow-Guided Frame Prediction](https://mlsanity.com/paper/2108.06852/) (2021)
- [Self-Supervised Predictive Convolutional Attentive Block for Anomaly Detection](https://mlsanity.com/paper/2111.09099/) (2021)
- [Anomaly Detection in Video Sequence with Appearance-Motion Correspondence](https://mlsanity.com/paper/1908.06351/) (2019)
- [Object-centric Auto-encoders and Dummy Anomalies for Abnormal Event Detection in Video](https://mlsanity.com/paper/1812.04960/) (2018)
- [Generative Cooperative Learning for Unsupervised Video Anomaly Detection](https://mlsanity.com/paper/2203.03962/) (2022)
- [Street Scene: A new dataset and evaluation protocol for video anomaly detection](https://mlsanity.com/paper/1902.05872/) (2019)
- [A Background-Agnostic Framework with Adversarial Training for Abnormal Event Detection in Video](https://mlsanity.com/paper/2008.12328/) (2020)
- [Learning Normal Dynamics in Videos with Meta Prototype Network](https://mlsanity.com/paper/2104.06689/) (2021)
- [Cloze Test Helps: Effective Video Anomaly Detection via Learning to Complete Video Events](https://mlsanity.com/paper/2008.11988/) (2020)
- [GODS: Generalized One-class Discriminative Subspaces for Anomaly Detection](https://mlsanity.com/paper/1908.05884/) (2019)
- [Video Anomaly Detection by Solving Decoupled Spatio-Temporal Jigsaw Puzzles](https://mlsanity.com/paper/2207.10172/) (2022)
- [Normalizing Flows for Human Pose Anomaly Detection](https://mlsanity.com/paper/2211.10946/) (2022)
- [SSMTL++: Revisiting Self-Supervised Multi-Task Learning for Video Anomaly Detection](https://mlsanity.com/paper/2207.08003/) (2022)
- [Multimodal Motion Conditioned Diffusion Model for Skeleton-based Video Anomaly Detection](https://mlsanity.com/paper/2307.07205/) (2023)
- [Multiscale Score Matching for Out-of-Distribution Detection](https://mlsanity.com/paper/2010.13132/) (2020)
- [Learning a distance function with a Siamese network to localize anomalies in videos](https://mlsanity.com/paper/2001.09189/) (2020)
- [Dynamic Local Aggregation Network with Adaptive Clusterer for Anomaly Detection](https://mlsanity.com/paper/2207.10948/) (2022)

## Built on by (most cited first, 22 of 22)

- [AnyAnomaly: Zero-Shot Customizable Video Anomaly Detection with LVLM](https://mlsanity.com/paper/2503.04504/) (2025)
- [A Unified Reasoning Framework for Holistic Zero-Shot Video Anomaly Analysis](https://mlsanity.com/paper/2511.00962/) (2025)
- [Track Any Anomalous Object: A Granular Video Anomaly Detection Pipeline](https://mlsanity.com/paper/2506.05175/) (2025)
- [No Need For Real Anomaly: MLLM Empowered Zero-Shot Video Anomaly Detection](https://mlsanity.com/paper/2602.19248/) (2026)
- [Sequential keypoint density estimator: an overlooked baseline of skeleton-based video anomaly detection](https://mlsanity.com/paper/2506.18368/) (2025)
- [Towards Adaptive Human-centric Video Anomaly Detection: A Comprehensive Framework and A New Benchmark](https://mlsanity.com/paper/2408.14329/) (2024)
- [A2Seek: Towards Reasoning-Centric Benchmark for Aerial Anomaly Understanding](https://mlsanity.com/paper/2505.21962/) (2025)
- [Pistachio: Towards Synthetic, Balanced, and Long-Form Video Anomaly Benchmarks](https://mlsanity.com/paper/2511.19474/) (2025)
- [Autoregressive Denoising Score Matching is a Good Video Anomaly Detector](https://mlsanity.com/paper/2506.23282/) (2025)
- [A Survey on Video Anomaly Detection via Deep Learning: Human, Vehicle, and Environment](https://mlsanity.com/paper/2508.14203/) (2025)
- [Detecting Contextual Anomalies by Discovering Consistent Spatial Regions](https://mlsanity.com/paper/2501.08470/) (2025)
- [Bounding-Box Trajectories Matter for Video Anomaly Detection](https://mlsanity.com/paper/2605.21957/) (2026)
- [Video Anomaly Detection with Structured Keywords](https://mlsanity.com/paper/2503.10653/) (2025)
- [Towards Scalable Video Anomaly Retrieval: A Synthetic Video-Text Benchmark](https://mlsanity.com/paper/2506.01466/) (2025)
- [Text-guided Fine-Grained Video Anomaly Understanding](https://mlsanity.com/paper/2511.00524/) (2025)
- [Action Hints: Semantic Typicality and Context Uniqueness for Generalizable Skeleton-based Video Anomaly Detection](https://mlsanity.com/paper/2509.11058/) (2025)
- [VADMamba++: Efficient Video Anomaly Detection via Hybrid Modeling in Grayscale Space](https://mlsanity.com/paper/2604.00360/) (2026)
- [Probe-VAD: Ordinal Likelihood Probing for Training-Free Video Anomaly Detection](https://mlsanity.com/paper/2609.17211/) (2026)
- [STEP: Score-Based Temporal Energy for Human Pose Video Anomaly Detection](https://mlsanity.com/paper/2608.19987/) (2026)
- [Look Inside Each Video: Rethinking How Video Anomaly Detection Is Evaluated](https://mlsanity.com/paper/2608.21854/) (2026)
- [NOVA: Normal-Side Modeling for Training-Free Zero-Shot Video Anomaly Detection](https://mlsanity.com/paper/2609.06360/) (2026)
- [Anomaly-Guided Observation and Two-Level Semantic Feedback for Training-Free Video Anomaly Analysis](https://mlsanity.com/paper/2610.16324/) (2026)

## Abstract (arXiv)

We propose a novel approach to video anomaly detection: we treat feature vectors extracted from videos as realizations of a random variable with a fixed distribution and model this distribution with a neural network. This lets us estimate the likelihood of test videos and detect video anomalies by thresholding the likelihood estimates. We train our video anomaly detector using a modification of denoising score matching, a method that injects training data with noise to facilitate modeling its distribution. To eliminate hyperparameter selection, we model the distribution of noisy video features across a range of noise levels and introduce a regularizer that tends to align the models for different levels of noise. At test time, we combine anomaly indications at multiple noise scales with a Gaussian mixture model. Running our video anomaly detector induces minimal delays as inference requires merely extracting the features and forward-propagating them through a shallow neural network and a Gaussian mixture model. Our experiments on five popular video anomaly detection benchmarks demonstrate state-of-the-art performance, both in the object-centric and in the frame-centric setup.