Mission

Training supervised surface anomaly detection models requires obtaining a large number anomalous and anomaly-free examples. While anomaly-free examples are often abundant, anomalous examples are rare and their visual properties may vary significantly, making the acquisition of a representative training set a difficult task. Visual anomaly detection methods are trained on anomaly-free examples only and focus on learning a visual model of the anomaly free appearance and detect deviations from the learned model as anomalies.

Our research focuses on the development of visual anomaly detection methods with an emphasis on the detection and localization of surface defects.

Projects

MV4.0Data-driven framework for development of machine vision solutions

October 2021 - September 2024

The functional objective of the project is to shift the paradigm in the development of machine vision solutions from hand-engineered specific solutions to data-driven learning-based design and development that would enable more general, efficient, flexible and economical development, deployment and maintenance of machine vision systems. The main research goal of this project is to develop novel deep learning methods for iterative, active, robust, weak, self-, unsupervised and few-shot learning that would reduce the amount of needed annotated data.

DIVIDDetection of inconsistencies in complex visual data using deep learning

July 2018 - December 2021

The objective of the project is to develop novel deep learning methods for modelling complex consistency and detecting inconsistencies in visual data using training images annotated with different levels of accuracy. The main project goal is to go beyond the traditional supervised learning, where all anomalies on all training images have to be adequately labelled.

Publications