Visual and Textual Recognition

Good perceptual capabilities are a key requirement for any intelligent agent. In particular, understanding the visual and textual world opens the door to mine two of the most common sources of information available today. In our group, we develop active research in areas such as:

i) Video analytics,

ii) Recommender systems,

iii) NLP technologies for mining documents, and

iv) Multimodal approaches that combine vision and text mining, among others.