Summary of Self-supervised Anomaly Detection in the Wild: Favor Joint Embeddings Methods, by Daniel Otero et al.
Self-Supervised Anomaly Detection in the Wild: Favor Joint Embeddings Methods
by Daniel Otero, Rafael Mateus, Randall Balestriero
First submitted to arxiv on: 5 Oct 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 investigates the application of Self-Supervised Learning (SSL) methods for real-world anomaly detection in sewer infrastructure inspection. It evaluates lightweight models like ViT-Tiny and ResNet-18 across various SSL frameworks, including BYOL, Barlow Twins, SimCLR, DINO, and MAE, under different class imbalance levels using the Sewer-ML dataset. The results show that joint-embedding methods like SimCLR and Barlow Twins outperform reconstruction-based approaches like MAE, particularly when dealing with class imbalance. Additionally, the study highlights the importance of model choice over backbone architecture and emphasizes the need for better label-free assessments of SSL representations. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper looks at how to use artificial intelligence (AI) to help detect problems in sewer systems without needing labeled data. The researchers tested different AI methods on a dataset of images from sewer inspections and found that some methods work better than others when dealing with uneven amounts of data between normal and abnormal samples. They also showed that the choice of AI model is more important than the type of computer chip used to run it. Overall, this study shows that AI can be useful for detecting problems in sewer systems without needing a lot of labeled training data. |
Keywords
» Artificial intelligence » Anomaly detection » Embedding » Mae » Resnet » Self supervised » Vit
