# Local Anomaly Detection in Videos using Object-Centric Adversarial Learning

arXiv: https://arxiv.org/abs/2011.06722 · HTML: https://mlsanity.com/paper/2011.06722/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Nov 2020.

## Built on by (most cited first, 7 of 7)

- [A Critical Study on the Recent Deep Learning Based Semi-Supervised Video Anomaly Detection Methods](https://mlsanity.com/paper/2111.01604/) (2021)
- [Multi-Task Learning based Video Anomaly Detection with Attention](https://mlsanity.com/paper/2210.07697/) (2022)
- [Human-Scene Network: A Novel Baseline with Self-rectifying Loss for Weakly supervised Video Anomaly Detection](https://mlsanity.com/paper/2301.07923/) (2023)
- [Object Class Aware Video Anomaly Detection through Image Translation](https://mlsanity.com/paper/2205.01706/) (2022)
- [Deep Crowd Anomaly Detection: State-of-the-Art, Challenges, and Future Research Directions](https://mlsanity.com/paper/2210.13927/) (2022)
- [Future Video Prediction from a Single Frame for Video Anomaly Detection](https://mlsanity.com/paper/2308.07783/) (2023)
- [Designing Multi-Robot Ground Video Sensemaking with Public Safety Professionals](https://mlsanity.com/paper/2602.08882/) (2026)

## Abstract (arXiv)

We propose a novel unsupervised approach based on a two-stage object-centric adversarial framework that only needs object regions for detecting frame-level local anomalies in videos. The first stage consists in learning the correspondence between the current appearance and past gradient images of objects in scenes deemed normal, allowing us to either generate the past gradient from current appearance or the reverse. The second stage extracts the partial reconstruction errors between real and generated images (appearance and past gradient) with normal object behaviour, and trains a discriminator in an adversarial fashion. In inference mode, we employ the trained image generators with the adversarially learned binary classifier for outputting region-level anomaly detection scores. We tested our method on four public benchmarks, UMN, UCSD, Avenue and ShanghaiTech and our proposed object-centric adversarial approach yields competitive or even superior results compared to state-of-the-art methods.