Summary of Softpatch: Unsupervised Anomaly Detection with Noisy Data, by Xi Jiang et al.
SoftPatch: Unsupervised Anomaly Detection with Noisy Data
by Xi Jiang, Ying Chen, Qiang Nie, Yong Liu, Jianlin Liu, Bin-Bin Gao, Jun Liu, Chengjie Wang, Feng Zheng
First submitted to arxiv on: 21 Mar 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a new unsupervised anomaly detection method called SoftPatch, designed to handle noisy data in real-world applications. The approach involves denoising image patches using noise discriminators before constructing coresets. This helps maintain strong modeling of normal data and avoids overconfidence issues. SoftPatch is tested on various noise scenarios and outperforms state-of-the-art methods on the MVTecAD and BTAD benchmarks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary SoftPatch is a new unsupervised anomaly detection method that can handle noisy data in real-world applications. The goal is to detect unusual images, but most current algorithms are not designed for this situation. SoftPatch tries to fix this problem by cleaning up the noise before looking for anomalies. It does this by using special tools called noise discriminators and a memory bank. This helps keep the model from getting too confident in its results. SoftPatch is tested on different types of noisy data and performs better than other methods on certain benchmarks. |
Keywords
* Artificial intelligence * Anomaly detection * Unsupervised
