5 papers
Enhancing Robustness in Post-Processing Watermarking: An Ensemble Attack Network Using CNNs and Transformers
Tzuhsuan Huang, Cheng Yu Yeo, Tsai-Ling Huang +3
Recent studies on deep watermarking have predominantly focused on in-processing watermarking, which integrates the watermarking process into image generation. However, post-process…
Invisible Backdoor Triggers in Image Editing Model via Deep Watermarking
Yu-Feng Chen, Tzuhsuan Huang, Pin-Yen Chiu +1
Diffusion models have achieved remarkable progress in both image generation and editing. However, recent studies have revealed their vulnerability to backdoor attacks, in which spe…
Gen-n-Val: Agentic Image Data Generation and Validation
Jing-En Huang, I-Sheng Fang, Tzuhsuan Huang +3
The data scarcity, label noise, and long-tailed category imbalance remain important and unresolved challenges in many computer vision tasks, such as object detection and instance s…
Adversarial Robustness for Unified Multi-Modal Encoders via Efficient Calibration
Chih-Ting Liao, Zhangquan Chen, Chunlei Meng +3
Recent unified multi-modal encoders align a wide range of modalities into a shared representation space, enabling diverse cross-modal tasks. Despite their impressive capabilities,…
BlenDA: Domain Adaptive Object Detection through diffusion-based blending
Tzuhsuan Huang, Chen-Che Huang, Chung-Hao Ku +1
Unsupervised domain adaptation (UDA) aims to transfer a model learned using labeled data from the source domain to unlabeled data in the target domain. To address the large domain…