Visual anomaly classification and segmentation are vital for automating
...
In practical scenarios where training data is limited, many predictive
s...
An energy-based model (EBM) is a popular generative framework that offer...
Any classifier can be "smoothed out" under Gaussian noise to build a new...
Test-time adaptation (TTA) is an emerging paradigm that addresses
distri...
Visual anomaly detection is commonly used in industrial quality inspecti...
Randomized smoothing is currently a state-of-the-art method to construct...
Recent works in Generative Adversarial Networks (GANs) are actively
revi...
Adversarial training (AT) is currently one of the most successful method...
Novelty detection, i.e., identifying whether a given sample is drawn fro...
A recent technique of randomized smoothing has shown that the worst-case...
In most real-world scenarios, labeled training datasets are highly
class...
Recent progress in deep convolutional neural networks (CNNs) have enable...