Research map / Label-efficient and robust learning / OOD and anomaly detection
Deep One-Class Classification via Interpolated Gaussian Descriptor
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use IGD - SSL instead of the strongest listed baseline | CIFAR10 anomaly detection · CIFAR10 | mean AUC: 0.8196 -> 0.9125 | Helped |
| Use IGD - SSL instead of the strongest listed baseline | MVTec AD detection · MVTec AD | AUC: 0.873 -> 0.934 | Helped |
| Use IGD - SSL instead of the strongest listed baseline | MVTec AD localisation · MVTec AD | pixel-level AUC: 0.89 -> 0.93 | Helped |
| Use IGD - SSL instead of the strongest listed baseline | Hyper-Kvasir anomaly detection · Hyper-Kvasir | AUC: 0.928 -> 0.937 | Helped |
| Use IGD - SSL instead of the strongest listed baseline | LAG anomaly detection · LAG | AUC: 0.819 -> 0.857 | Helped |
| Add MS-SSIM + MAE reconstruction losses rather than MSE (ablation) | CIFAR10 anomaly detector · CIFAR10 | AUC - Full: 0.615 -> 0.731 | Helped |
| Add Gaussian anomaly classifier to REC (ablation) | CIFAR10 anomaly detector · CIFAR10 | AUC - Full: 0.731 -> 0.819 | Helped |
| Add adversarial interpolation regularisation to REC + GAC (ablation) | CIFAR10 anomaly detector · CIFAR10 | AUC - Full: 0.819 -> 0.836 | Helped |
| Add MS-SSIM + MAE reconstruction losses rather than MSE (ablation) | CIFAR10 anomaly detector · CIFAR10 (20% training data) | AUC: 0.552 -> 0.655 | Helped |
| Add Gaussian anomaly classifier to REC (ablation) | CIFAR10 anomaly detector · CIFAR10 (20% training data) | AUC: 0.655 -> 0.785 | Helped |
| Add adversarial interpolation regularisation to REC + GAC (ablation) | CIFAR10 anomaly detector · CIFAR10 (20% training data) | AUC: 0.785 -> 0.822 | Helped |
| Add Gaussian anomaly classifier to REC (ablation) | CIFAR10 anomaly detector · CIFAR10 (10% contamination) | AUC: 0.677 -> 0.781 | Helped |
Closest papers by meaning
- OCGAN: One-Class Novelty Detection Using GANs With Constrained Latent Representations (CVPR 2019 · similarity 0.87)
- Adversarially Learned One-Class Classifier for Novelty Detection (CVPR 2018 · similarity 0.86)
- A Simple Unified Framework for Detecting Out-of-Distribution Samples and Adversarial Attacks (NeurIPS 2018 · similarity 0.86)
- Learning and Evaluating Representations for Deep One-Class Classification (ICLR 2021 · similarity 0.86)
- CASE: Exploiting Intra-class Compactness and Inter-class Separability of Feature Embeddings for Out-of-Distribution Detection (AAAI 2024 · similarity 0.85)
- ResAD: A Simple Framework for Class Generalizable Anomaly Detection (NeurIPS 2024 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.