Assisting people in efficiently producing visually plausible 3D characte...
Real world data often exhibits a long-tailed and open-ended (with unseen...
We present Masked Frequency Modeling (MFM), a unified frequency-domain-b...
Motion style transfer is a common method for enriching character animati...
Contrastive self-supervised learning has largely narrowed the gap to
sup...
Unsupervised contrastive learning achieves great success in learning ima...
Recent unsupervised contrastive representation learning follows a Single...
Contrastive learning has recently shown immense potential in unsupervise...
Joint clustering and feature learning methods have shown remarkable
perf...
Natural scene understanding is a challenging task, particularly when
enc...
Face clustering is an essential tool for exploiting the unlabeled face d...
Learning a good image prior is a long-term goal for image restoration an...
Existing works on domain adaptation often assume clear boundaries betwee...
Normalization methods are essential components in convolutional neural
n...
Real world data often have a long-tailed and open-ended distribution. A
...
Face recognition sees remarkable progress in recent years, and its
perfo...
Intelligent agent naturally learns from motion. Various self-supervised
...
Face recognition has witnessed great progress in recent years, mainly
at...
Deep convolutional networks for semantic image segmentation typically re...