S
Shizhe Hu
Researcher at Zhengzhou University
Publications - 22
Citations - 245
Shizhe Hu is an academic researcher from Zhengzhou University. The author has contributed to research in topics: Cluster analysis & Information bottleneck method. The author has an hindex of 5, co-authored 15 publications receiving 79 citations.
Papers
More filters
Journal ArticleDOI
Deep multi-view learning methods: a review
TL;DR: In this article, a comprehensive review on deep multi-view learning from the following two perspectives: MVL methods in deep learning scope and deep MVL extensions of traditional methods is presented, and the authors attempt to identify some open challenges to inform future research directions.
Journal ArticleDOI
DSPNet: Deep scale purifier network for dense crowd counting
TL;DR: A novel deep scale purifier network (DSPNet) that can encode multiscale features and reduce the loss of contextual information for dense crowd counting and is end-to-end and has a fully convolutional architecture.
Proceedings ArticleDOI
Multi-task Clustering of Human Actions by Sharing Information
TL;DR: This work presents a novel and effective Multi-Task Information Bottleneck (MTIB) clustering method, which is capable of exploring the shared information between multiple action clustering tasks to improve the performance of individual task.
Journal ArticleDOI
DMIB: Dual-Correlated Multivariate Information Bottleneck for Multiview Clustering.
TL;DR: A novel dual-correlated multivariate information bottleneck (DMIB) method for MVC that is able to explore both interfeature correlations and intercluster correlations and theoretically prove the convergence of the proposed algorithm.
Journal ArticleDOI
Joint specific and correlated information exploration for multi-view action clustering
TL;DR: This work designs a new Bag-of-Shared-Words (BoSW) model to discover the view-shared visual words that preserve the consistency among visual words of different views and presents a novel JOint INformation boTtleneck (JOINT) algorithm, which compresses the actions of each view while jointly preserving the complementary view-specific information and correlated information among views.