FASTensor: A tensor framework for spatiotemporal description

Abstract

Spatiotemporal description is a research field with applications in various areas such as video indexing, surveillance, human-computer interfaces, among others. Big Data problems in large databases are now being treated with Deep Learning tools, however we still have room for improvement in spatiotemporal handcraft description. Moreover, we still have problems that involve small data in which data augmentation and other techniques are not valid. The main contribution of this Ph.D. Thesis 1 is the development of a framework for spatiotemporal representation using orientation tensors enabling dimension reduction and invariance. This is a multipurpose framework called Features As Spatiotemporal Tensors (FASTensor). We evaluate this framework in three different applications: Human Action recognition, Video Pornography classification and Cancer Cell classification. The latter one is also a contribution of this work, since we introduce a new dataset called Melanoma Cancer Cell dataset (MCC). It is a small data that cannot be artificially augmented due the difficulty of extraction and the nature of motion. The results were competitive, while also being fast and simple to implement. Finally, our results in the MCC dataset can be used in other cancer cell treatment analysis.