Tantárgy adatlapja
The group of artificial intelligence methods named "Deep Learning" show excellent performances in computer vision tasks, such as scene understanding, object recognition in images and video and action recognition in video. The main objective of this course is that students learn to design and implement a Deep Learning approach for visual scenes understanding. They will get the following skills: Design and implement an adequate data extraction and augmentation strategy; Set-up an architecture using known models; Select and parameterize optimisation algorithms; Design transfer-learning and information fusion schemes; Implement Deep Learning algorithms wit open source frameworks; Evaluate performances of the designed approaches in terms of common quality metrics.
