Main navigation
- Programs
- Subjects
- Universities
- Destinations
- Advice
Global research into developing intelligent machines is progressing at a remarkable pace. Our team actively participates in this advancement by investigating how to merge breakthroughs in deep learning, symbolic reasoning, computer vision, natural language processing, and robotics. This encompasses developing systems that synthesize information from diverse inputs like video, text, images, audio, and satellite data. Additionally, we focus on methods that minimize or eliminate the need for supervision when learning from raw data, such as deriving word meanings by analyzing video caption text.
Supporting this comprehensive approach, we conduct foundational theoretical studies in qualitative spatial reasoning, expanding on our groundbreaking Region Connection Calculus (RCC) framework. Our work also includes developing algorithms for human pose detection, automated text alignment and segmentation, user behavior modeling and customization, language analysis using text corpora, and robotics applications, with special emphasis on manipulation techniques and path planning.