Xiaoqian Qin

TL;DR: The proposed method outperforms several previous state of the art methods, while could also be used to significantly boost the performance of one-versus-one kinship verification when the information about both parents are available.

TL;DR: In this paper, a relative symmetric bilinear model (RSBM) is introduced to model the similarity between the child and the parents, by incorporating the prior knowledge that a child may resemble one particular parent more than the other.

TL;DR: Most of the participants tackled the image-restricted challenge and experimental results demonstrated better kinship verification performance than the baseline methods provided by the organizers.

TL;DR: This work introduces a new type of learning problem, called mixed bi-subject kinship verification, and proposes a novel multi-task learning method to address this problem with two transformation matrices - one is shared amongst all the tasks and the other is unique to each task.

TL;DR: This paper attaches great importance to the difficulties in practical applications of kinship verification, and reviews the prominent algorithms from the perspective of learning more efficient models with more diverse kin relations, and shows how to develop an efficient and robust kinships verification system.