Long-term object detection requires the integration of frame-based results over several
seconds. For non-deformable objects, long-term detection is often addressed using object …

We study the problem of detecting talking activities in collaborative learning videos. Our
approach uses head detection and projections of the log-magnitude of optical flow vectors to …

There is a growing need for developing fast and scalable methods for storing and
processing large databases of healthcare images and videos. The paper reviews current …

Human activity recognition in videos is a challenging problem that has drawn a lot of
interest, particularly when the goal requires the analysis of a large video database. The …

The dissertation develops new methods for assessing student participation in long (> 1 hour)
classroom videos. First, the dissertation introduces the use of multiple image representations …

Audio recordings of collaborative learning environments contain a constant presence of
cross-talk and background noise. Dynamic speech recognition between Spanish and …

Mathematics is a topic very important to achieve success during education and for life. In this
way, the factors that could influence on mathematics learning, from the point of the view of …

Speaker identification in noisy audio recordings, specifically those from collaborative
learning environments, can be extremely challenging. There is a need to identify individual …

Convolutional Neural Networks (CNN) have provided new and accurate methods for
processing digital images and videos. Yet, training CNNs is extremely demanding in terms …

Classification of human activity in raw video presents a challenging problem that remains
unsolved, and is of great interest for large datasets. Though there have been several …