# Memory Efficient Continual Learning for Edge-Based Visual Anomaly Detection

arXiv: https://arxiv.org/abs/2503.02691 · HTML: https://mlsanity.com/paper/2503.02691/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 26 references that are in the map.
- First posted on arXiv 4 Mar 2025.

## Builds on (most cited first, 19 of 19)

- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Learning without Forgetting](https://mlsanity.com/paper/1606.09282/) (2016)
- [Progressive Neural Networks](https://mlsanity.com/paper/1606.04671/) (2016)
- [Experience Replay for Continual Learning](https://mlsanity.com/paper/1811.11682/) (2018)
- [PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruning](https://mlsanity.com/paper/1711.05769/) (2017)
- [PathNet: Evolution Channels Gradient Descent in Super Neural Networks](https://mlsanity.com/paper/1701.08734/) (2017)
- [MCUNet: Tiny Deep Learning on IoT Devices](https://mlsanity.com/paper/2007.10319/) (2020)
- [SegmentMeIfYouCan: A Benchmark for Anomaly Segmentation](https://mlsanity.com/paper/2104.14812/) (2021)
- [Student-Teacher Feature Pyramid Matching for Anomaly Detection](https://mlsanity.com/paper/2103.04257/) (2021)
- [Latent Replay for Real-Time Continual Learning](https://mlsanity.com/paper/1912.01100/) (2019)
- [Rethinking Experience Replay: a Bag of Tricks for Continual Learning](https://mlsanity.com/paper/2010.05595/) (2020)
- [BMAD: Benchmarks for Medical Anomaly Detection](https://mlsanity.com/paper/2306.11876/) (2023)
- [Imbalanced Continual Learning with Partitioning Reservoir Sampling](https://mlsanity.com/paper/2009.03632/) (2020)
- [IM-IAD: Industrial Image Anomaly Detection Benchmark in Manufacturing](https://mlsanity.com/paper/2301.13359/) (2023)
- [PaSTe: Improving the Efficiency of Visual Anomaly Detection at the Edge](https://mlsanity.com/paper/2410.11591/) (2024)
- [Unveiling the Anomalies in an Ever-Changing World: A Benchmark for Pixel-Level Anomaly Detection in Continual Learning](https://mlsanity.com/paper/2403.15463/) (2024)
- [Continual Learning Approaches for Anomaly Detection](https://mlsanity.com/paper/2212.11192/) (2022)
- [A Benchmark and Empirical Analysis for Replay Strategies in Continual Learning](https://mlsanity.com/paper/2208.02660/) (2022)
- [Improving unsupervised anomaly localization by applying multi-scale memories to autoencoders](https://mlsanity.com/paper/2012.11113/) (2020)

## Built on by (most cited first, 4 of 4)

- [Towards Continual Visual Anomaly Detection in the Medical Domain](https://mlsanity.com/paper/2508.18013/) (2025)
- [Continual Visual Anomaly Detection on the Edge: Benchmark and Efficient Solutions](https://mlsanity.com/paper/2604.06435/) (2026)
- [MoViAD: A Modular Library for Visual Anomaly Detection](https://mlsanity.com/paper/2507.12049/) (2025)
- [Rethinking Continual Anomaly Detection on the Edge: Benchmarking Under Realistic Industrial Conditions](https://mlsanity.com/paper/2605.24251/) (2026)

## Abstract (arXiv)

Visual Anomaly Detection (VAD) is a critical task in computer vision with numerous real-world applications. However, deploying these models on edge devices presents significant challenges, such as constrained computational and memory resources. Additionally, dynamic data distributions in real-world settings necessitate continuous model adaptation, further complicating deployment under limited resources. To address these challenges, we present a novel investigation into the problem of Continual Learning for Visual Anomaly Detection (CLAD) on edge devices. We evaluate the STFPM approach, given its low memory footprint on edge devices, which demonstrates good performance when combined with the Replay approach. Furthermore, we propose to study the behavior of a recently proposed approach, PaSTe, specifically designed for the edge but not yet explored in the Continual Learning context. Our results show that PaSTe is not only a lighter version of STPFM, but it also achieves superior anomaly detection performance, improving the f1 pixel performance by 10% with the Replay technique. In particular, the structure of PaSTe allows us to test it using a series of Compressed Replay techniques, reducing memory overhead by a maximum of 91.5% compared to the traditional Replay for STFPM. Our study proves the feasibility of deploying VAD models that adapt and learn incrementally on CLAD scenarios on resource-constrained edge devices.