A Weak Supervision Approach to Detecting Visual Anomalies for Automated Testing of Graphics Units
This work addresses the challenge of automated testing for graphics units, offering a more efficient solution for hardware developers, though it is incremental in improving upon existing weak supervision techniques.
The paper tackles the problem of detecting visual anomalies in graphics unit testing by introducing a weakly supervised deep learning system that eliminates the need for manual labeling. The method significantly outperforms unsupervised baselines and discovers previously undetected corruptions while meeting real-time accuracy and efficiency requirements.
We present a deep learning system for testing graphics units by detecting novel visual corruptions in videos. Unlike previous work in which manual tagging was required to collect labeled training data, our weak supervision method is fully automatic and needs no human labelling. This is achieved by reproducing driver bugs that increase the probability of generating corruptions, and by making use of ideas and methods from the Multiple Instance Learning (MIL) setting. In our experiments, we significantly outperform unsupervised methods such as GAN-based models and discover novel corruptions undetected by baselines, while adhering to strict requirements on accuracy and efficiency of our real-time system.