The rapid development of facial manipulation techniques has aroused public concerns in recent years. Existing deepfake video detection approaches attempt to capture the discrim- inative features between real and fake faces based on tem- poral modelling. However, these works impose supervisions sparsely sampled frames but overlook local mo- tions among adjacent frames, which instead encode rich ...

Intrinsic imaging or intrinsic image decomposition has traditionally been described as the problem of decomposing an into two layers: a reflectance, albedo invariant color material; and shading, produced by interaction between light geometry. Deep learning techniques have broadly applied in recent years to increase accuracy those separations. In this survey, we overview results context well-kno...

We describe an algorithm for automatically segmenting flowers in colour photographs. This is a challenging problem because of the sheer variety of flower classes, the variability within a class and within a particular flower, and the variability of the imaging conditions – lighting, pose, foreshortening, etc. The method couples two models – a colour model for foreground and background, and a li...

Abstract Purpose: With the rising incidence of early-onset pancreatic cancer (EOPC), molecular characteristics that distinguish ductal adenocarcinoma (PDAC) tumors from those arising at a later age are not well understood. Experimental Design: We performed bioinformatic analysis genomic and transcriptomic data generated 269 advanced (metastatic or locally advanced) 277 resectable PDAC tumor sam...