Transformaly – Two (Feature Spaces) Are Better Than One

by   Matan Jacob Cohen, et al.

Anomaly detection is a well-established research area that seeks to identify samples outside of a predetermined distribution. An anomaly detection pipeline is comprised of two main stages: (1) feature extraction and (2) normality score assignment. Recent papers used pre-trained networks for feature extraction achieving state-of-the-art results. However, the use of pre-trained networks does not fully-utilize the normal samples that are available at train time. This paper suggests taking advantage of this information by using teacher-student training. In our setting, a pretrained teacher network is used to train a student network on the normal training samples. Since the student network is trained only on normal samples, it is expected to deviate from the teacher network in abnormal cases. This difference can serve as a complementary representation to the pre-trained feature vector. Our method – Transformaly – exploits a pre-trained Vision Transformer (ViT) to extract both feature vectors: the pre-trained (agnostic) features and the teacher-student (fine-tuned) features. We report state-of-the-art AUROC results in both the common unimodal setting, where one class is considered normal and the rest are considered abnormal, and the multimodal setting, where all classes but one are considered normal, and just one class is considered abnormal. The code is available at


Teacher-Student Network for 3D Point Cloud Anomaly Detection with Few Normal Samples

Anomaly detection, which is a critical and popular topic in computer vis...

DeSTSeg: Segmentation Guided Denoising Student-Teacher for Anomaly Detection

Visual anomaly detection, an important problem in computer vision, is us...

CFA: Coupled-hypersphere-based Feature Adaptation for Target-Oriented Anomaly Localization

For a long time, anomaly localization has been widely used in industries...

CA2: Class-Agnostic Adaptive Feature Adaptation for One-class Classification

One-class classification (OCC), i.e., identifying whether an example bel...

Complementary Pseudo Multimodal Feature for Point Cloud Anomaly Detection

Point cloud (PCD) anomaly detection steadily emerges as a promising rese...

Interpreting A Pre-trained Model Is A Key For Model Architecture Optimization: A Case Study On Wav2Vec 2.0

A deep Transformer model with good evaluation score does not mean each s...

CBIR using Pre-Trained Neural Networks

Much of the recent research work in image retrieval, has been focused ar...

Please sign up or login with your details

Forgot password? Click here to reset