arXiv (Cornell University) · 2020 · 109 citations · 17 references
ComputingComputer ScienceComputer VisionFeature EngineeringMachine LearningAnomaly DetectionDeep LearningFeature DetectionRepresentation LearningSelf-supervised LearningSupervised LearningDeep Learning-based Anomaly DetectionImage Anomaly DetectionNoise RobustnessRobustness To NoiseSelf-supervised PretrainingTrainingTraining Time
Nearest neighbors is a successful and long-standing technique for anomaly detection. Significant progress has been recently achieved by self-supervised deep methods (e.g. RotNet). Self-supervised features however typically under-perform Imagenet pre-trained features. In this work, we investigate whether the recent progress can indeed outperform nearest-neighbor methods operating on an Imagenet pretrained feature space. The simple nearest-neighbor based-approach is experimentally shown to outperform self-supervised methods in: accuracy, few shot generalization, training time and noise robustness while making fewer assumptions on image distributions.
17
Image quality assessment: from error visibility to structural similarity
Zhou Wang, Alan C. Bovik, Hamid R. Sheikh et al. · IEEE Transactions on Image Processing · 2004 · 56K citations
Robust principal component analysis?
Emmanuel J. Candès, Xiaodong Li, Yi Ma et al. · Journal of the ACM · 2011 · 6.9K citations
Caltech-UCSD Birds-200-2011 Dataset
Catherine Wah · CaltechAUTHORS (California Institute of Technology) · 2024 · 3.9K citations · Full text