Deep Learning has emerged as the dominant machine learning paradigm, leading to state-of-the-art performance in numerous application domains involving learning directly from data. The power of Deep Learning models comes from their ability to receive as input raw data and jointly optimize multiple data representations of increasing level of abstraction towards achieving the overall (regression/ classification/clustering) goal. However, this brute-force way of training deep network architectures using data sets of enormous size creates new methodological and practical challenges that need to be properly addressed. This Special Issue seeks for new contributions in Deep Learning models and their applications in Visual Content Analysis problems.
Topics of interest to the Special Issue include (but are not limited to):
Representation learning for visual data clustering, classification and analysis
Unsupervised and semi-supervised methods for media analysis based on deep learning
Deep learning for data indexing and retrieval
Multi-view and cross-view Deep Learning based visual content analysis
Embedded Deep Learning systems for visual content analysis
Deep learning for 3D Computer Vision
Deep visual analysis for autonomous systems
Explainable deep learning for visual data analysis
Semantic visual analysis: human activity recognition, face/facial expression recognition, scene understanding, object detection and tracking, saliency detection
Multimedia data (signal, 2D/3D image, video) analysis in medicine, science and engineering