People
Researchers, assistants, and students building visual intelligence across computer vision, trustworthy AI, generative models, physical AI, and multimodal agents.
Research Center for Information Technology Innovation, Academia Sinica
Computer vision, machine learning, deep learning, biometrics, face recognition, and visual surveillance.
Generative models, self/weakly-supervised learning, depth estimation, style transfer, computer vision, and creative AI.
Trustworthy AI, adversarial robustness, privacy, semantic watermark security, and reliable generative AI.
Diffusion-based object detection, watermark robustness, deepfake detection, and AI security.
Image editing security, invisible backdoor triggers, deep watermarking, and generative-model robustness.
Multimodal large language models, hallucination subspaces, and reliable MLLM estimation.
Computer vision and multimodal AI research.
Weather foundation model and embodied AI.
Weather foundation model.