Human-computer Interaction Framework project
Current state-of-the-art human-computer interaction platform still focuses on voice/text-based interaction. The goal of this project to build the next-generation human-computer interaction framework that empowers visually grounded interactions for physical robots or virual agents.
Visual supervision evaluation platform
Current computer vision (CV) systems are evalauted in controlled environment such as object detection or segmentation. The goal of this project is to create a benchmark that evalaute systems' robustness and accuracy in real-world scenairos, which in turn foster future research for real-world CV systems.
Multimodal public opinion understanding algorithm
Multimodal understanding is becoming increasingly important for social media trend analyisis. The goal of thi sproject to develop algorithms and computation models that better parse and predict event trends and build event knowledge base.