Shiguang Shan

TL;DR: A novel non-statistics based face representation approach, local Gabor binary pattern histogram sequence (LGBPHS), in which training procedure is unnecessary to construct the face model, so that the generalizability problem is naturally avoided.

TL;DR: Experimental results on the Brodatz and KTH-TIPS2-a texture databases show that WLD impressively outperforms the other widely used descriptors (e.g., Gabor and SIFT), and experimental results on human face detection also show a promising performance comparable to the best known results onThe MIT+CMU frontal face test set, the AR face data set, and the CMU profile test set.

TL;DR: The evaluation protocol based on the CAS-PEAL-R1 database is discussed and the performance of four algorithms are presented as a baseline to do the following: elementarily assess the difficulty of the database for face recognition algorithms; preference evaluation results for researchers using the database; and identify the strengths and weaknesses of the commonly used algorithms.

TL;DR: A novel Deep Supervised Hashing method to learn compact similarity-preserving binary code for the huge body of image data and extensive experiments show the promising performance of the method compared with the state-of-the-arts.

TL;DR: The proposed method is extended for attribute style manipulation in an unsupervised manner and outperforms the state-of-the-art on realistic attribute editing with other facial details well preserved.