Simone Bonechi (DIISM, University of Siena)
March 1, 2018 – 9:30 AM
DIISM, Artificial Intelligence laboratory (room 201), Siena SI
Semantic Image segmentation is more and more being of interest for computer vision and machine learning researchers. A lot of applications need accurate and efficient segmentation mechanisms: autonomous driving, indoor navigation, and even virtual or augmented reality systems to name a few. This demand coincides with the rise of deep learning approaches in almost every field or application target related to computer vision. The aim of this seminar is firstly to introduce the main background concepts of semantic image segmentation. Then some state of the art convolutional neural networks used in this field will be presented describing their structure and showing the obtained results on standard datasets. The last part of the seminar will introduce some promising works, in which I’m currently involved, that will be deepened in future presentations.