Xiaofei Xie

TL;DR: A novel adversarial attack method that can generate visually natural motion-blurred adversarial examples, named motion-based adversarial blur attack (ABBA), which shows more effective penetrating capability to the state-of-the-art GAN-based deblurring mechanisms compared with other blurring methods.

TL;DR: A large-scale study into the capability of multiple uncertainty metrics in differentiating benign examples (BEs) and AEs, which enables to characterize the uncertainty patterns of input data and proposes an automated testing technique to generate multiple types of uncommon AEs and BEs that are largely missed by existing techniques.

TL;DR: This work investigates and introduces a new type of adversarial attack to evade FR systems by manipulating facial content, called adversarial morphing attack (a.k.a. Amora), and indicates that a novel black-box adversarial attacked based on local deformation is possible, and is vastly different from additive noise attacks.

TL;DR: A model-free deraining method, EfficientDeRain, which is able to process a rainy image within 10~ms, over 80 times faster than the state-of-the-art method (i.e., RCDNet), while achieving similar de-rain effects, and an effective data augmentation method that helps to train network for real rainy image handling.

TL;DR: An in-depth evaluation on a state-of-the-art speech-to-text DL system demonstrates the effectiveness of the model RNN as an abstract state transition system in improving quality and reliability of stateful DL systems.