The field of speech processing has undergone a transformative shift with the advent of deep
learning. The use of multiple processing layers has enabled the creation of models capable …

Advances in computing technology have fostered the development of new and powerful
deep learning (DL) techniques, which have demonstrated promising results in a wide range …

In this work, we tackle the challenging problem of arbitrary image style transfer using a novel
style feature representation learning method. A suitable style representation, as a key …

We consider the problem of obtaining image quality representations in a self-supervised
manner. We use prediction of distortion type and degree as an auxiliary task to learn …

Supervised crowd counting relies heavily on costly manual labeling, which is difficult and
expensive, especially in dense scenes. To alleviate the problem, we propose a novel …

Comparative settings (eg. pairwise choice, listwise ranking) have been adopted by a wide
range of subjective studies for image quality assessment (IQA), as it inherently standardizes …

Semi-supervised learning, ie jointly learning from labeled and unlabeled samples, is an
active research topic due to its key role on relaxing human supervision. In the context of …

Learning time-series representations when only unlabeled data or few labeled samples are
available can be a challenging task. Recently, contrastive self-supervised learning has …

The mainstream crowd counting methods usually utilize the convolution neural network
(CNN) to regress a density map, requiring point-level annotations. However, annotating …

In the last decade, crowd counting and localization attract much attention of researchers due
to its wide-spread applications, including crowd monitoring, public safety, space design, etc …