Panel on Applied Deep Learning in Satellite Images Using Small Supervised Data
Alessandro Nicolosi
-
Members: FreeCIS
IEEE Members: Free
Non-members: FreeLength: 01:11:24
20 Jul 2022
The need to have annotated data to train a model, is one of the
weaknesses of the technologies based on Deep Learning, especially in
operational scenarios where it is not possible to annotate data in a
reasonable time frame or in cases where you do not have enough data
available related to rare scenarios. The purpose of this panel deals
with issues related to methodologies for training neural networks with
little data, using Self Supervised Learning techniques.
weaknesses of the technologies based on Deep Learning, especially in
operational scenarios where it is not possible to annotate data in a
reasonable time frame or in cases where you do not have enough data
available related to rare scenarios. The purpose of this panel deals
with issues related to methodologies for training neural networks with
little data, using Self Supervised Learning techniques.
