---
tags: 論文
---

# 論文、理論收集 Anomaly Detection

- 監督式、傳統ML
- https://hackmd.io/pcl8XUuVRlufm5e-zv7Ikw
- f-AnoGAN: Fast unsupervised anomaly detection with generative adversarial networks
- https://hackmd.io/zFzy4Kz-StGppPJoXFvwvg
- Multiresolution Knowledge Distillation for Anomaly Detection
- https://hackmd.io/IvZ-I023Qxu_5e8ptQcgDQ
- https://github.com/Niousha12/Knowledge_Distillation_AD
- 實驗: class會影響到正確率，有些物件好辨識，有些難
- GAN Ensemble for Anomaly Detection
- https://hackmd.io/zFzy4Kz-StGppPJoXFvwvg
- https://github.com/tufts-ml/GAN-Ensemble-for-Anomaly-Detection
- Self-trained Deep Ordinal Regression for End-to-End Video Anomaly Detection
- https://arxiv.org/abs/2003.06780
- 演算法自我挑選anomalous frame normal frame
- 最初訓練模型時anomalous frame最好是其他影片的影像
- Same Same But DifferNet: Semi-Supervised Defect Detection with Normalizing Flows
- https://hackmd.io/i0qPSoBvS_mko7PcDhk9Kw
- https://arxiv.org/abs/2008.12577
- PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localization (ICPR 2020)
- https://hackmd.io/4fepGO6CQnyzRfpgXPXBFA

## noisy label (帶噪學習)
- Confident Learning: Estimating Uncertainty in Dataset Labels
- https://hackmd.io/2wTXgRYsQryXKAqZ2ZqB_A
- https://github.com/cgnorthcutt/cleanlab

## anomaly detection transfer learning