Y
Yu Kong
Researcher at Northeastern University
Publications - 44
Citations - 4642
Yu Kong is an academic researcher from Northeastern University. The author has contributed to research in topics: Feature extraction & Discriminative model. The author has an hindex of 20, co-authored 42 publications receiving 3030 citations. Previous affiliations of Yu Kong include Beijing Institute of Technology & State University of New York System.
Papers
More filters
Proceedings ArticleDOI
Residual Dense Network for Image Super-Resolution
TL;DR: This paper proposes residual dense block (RDB) to extract abundant local features via dense connected convolutional layers and uses global feature fusion in RDB to jointly and adaptively learn global hierarchical features in a holistic way.
Posted Content
Residual Dense Network for Image Super-Resolution
TL;DR: Zhang et al. as mentioned in this paper proposed a residual dense block (RDB) to extract abundant local features via dense connected convolutional layers, which further allows direct connections from the state of preceding RDB to all the layers of current RDB, leading to a contiguous memory mechanism.
Book ChapterDOI
A Discriminative Model with Multiple Temporal Scales for Action Prediction
Yu Kong,Dmitry Kit,Yun Fu +2 more
TL;DR: A novel discriminative multi-scale model for predicting the action class from a partially observed video, which captures temporal dynamics of human actions by explicitly considering all the history of observed features as well as features in smaller temporal segments.
Proceedings ArticleDOI
Deep Sequential Context Networks for Action Prediction
Yu Kong,Zhiqiang Tao,Yun Fu +2 more
TL;DR: This paper proposes efficient and powerful deep networks for action prediction from partially observed videos containing temporally incomplete action executions, and develops a new learning formulation that enables efficient model training.
Book ChapterDOI
Learning human interaction by interactive phrases
Yu Kong,Yunde Jia,Yun Fu +2 more
TL;DR: A novel hierarchical model to encode interactive phrases based on the latent SVM framework where interactive phrases are treated as latent variables is proposed to deal with motion ambiguity and partial occlusion in interactions.