You are here

Deep Learning

Main activities

Recognizing and classifying all the entities appearing in an image is a fundamental goal of computer vision, and constitutes a key element in the semantic understanding of images, videos and other multi-media resources. This macro-activity is concerned with the development of deep learning theory and practice to understand the image content. More images depicting the same objects can be used for a 3D reconstruction of the scene.

Non-invasive technologies for monitoring and understanding complex environments have applications in diverse domains such as security in industrial environment, outdoor and indoor surveillance, person action recognition, traffic analysis, assisted living, customer behaviour, sports analysis etc. This macro-activity is concerned with the development of computer vision technologies for dynamic scene understanding in such applicative contexts. The most recently used techniques are based on machine learning and deep learning with structured output prediction.

We investigate how to deploy representation learning within the framework of decision forests, which are ensembles of binary decision trees that have become very popular in computer vision.



If a piece of valuable wood has some defects, it cannot be used to make certain objects, but some others can be made avoiding the defect zone.

SpinRetail is a project coordinated by Spindox Labs for the implementation of a smart system which aims to monitor and analyse users behaviour within retail outlets.