Ad

awesome-anomaly-detection: A curated list of anomaly detection resources

Anomaly detection resources: Find survey papers, articles, code, and more for time-series and video-level anomaly detection.
Screenshot of hoya012/awesome-anomaly-detection homepage

Anomaly detection identifies unusual patterns from expected behavior, treating it as an unsupervised learning problem. This repository provides a curated list of resources, including survey papers and articles, to aid in exploring various techniques and approaches in anomaly detection. It covers time-series and video-level anomaly detection, offering a comprehensive overview of the field.

This repository offers a focused collection of resources, balancing survey papers with practical implementation details. It specifically caters to both time-series and video anomaly detection, which are distinct but important areas. The inclusion of code links facilitates direct exploration and experimentation with discussed methods.

  • Time-series detection: Resources for identifying anomalies in time-series data, including LSTM networks and deep learning approaches.
  • Video-level detection: Resources for anomaly detection in videos, including spatiotemporal autoencoders and deep learning models.
  • Survey papers: Comprehensive reviews of different anomaly detection techniques and deep learning methods.
  • Procceedings: Links to relevant conference proceedings like CVPR, ICCV and NeurIPS focusing on cutting-edge research.
  • Code implementations: Links to GitHub repositories offering code for various anomaly detection algorithms.
  • Applications: Resources covering real-world applications such as intrusion detection and surveillance.
  • Theoretical Foundations: Provides links to foundational papers and reviews in anomaly detection.

The repository is actively maintained with recent updates and additions. The inclusion of resources from 2010 to 2022 signifies ongoing research and development in the field. The variety of source types indicates a comprehensive effort to gather relevant materials. The constant addition of references to new papers is clear indicator of continued support.

This resource is valuable for researchers, developers, and anyone interested in anomaly detection. It provides a structured way to discover relevant literature, code, and applications, supporting both theoretical understanding and practical implementation in time-series and video analysis.

Summarize:
Share:
Stars
2,903
Forks
510
Issues
9
Created
8 years ago
Commit
3 years ago
License
None
Archived
No
Updated 1 day ago

Similar Repositories