In So Kweon

Latest

• Towards Robust Data Hiding against (JPEG) Compression: A Pseudo-differentiable Deep Learning Approach
• Universal Adversarial Perturbations Through the Lens of Deep Steganography: A Fourier Perspective
• UDH: Universal Deep Hiding for Steganography, Watermarking, and Light Field Messaging
• Robustness May Be at Odds with Fairness: An Empirical Study on Class-wise Accuracy
• ResNet or DenseNet? Introducing Dense Shortcuts to ResNet
• Batch Normalization Increases Adversarial Vulnerability: Disentangling Usefulness and Robustness of Model Features
• Revisiting Batch Normalization for Improving Corruption Robustness
• Double Targeted Universal Adversarial Perturbations
• Understanding Adversarial Examples from the Mutual Influence of Images and Perturbations
• Data from Model: Extracting Data from Non-robust and Robust Models
• CD-UAP: Class Discriminative Universal Adversarial Perturbation
• Propose-and-Attend Single Shot Detector
• Revisiting Residual Networks with Nonlinear Shortcuts
• Fast Perception, Planning, and Execution for a Robotic Butler: Wheeled Humanoid M-Hubo