C
Ce Zhu
Researcher at University of Electronic Science and Technology of China
Publications - 319
Citations - 8042
Ce Zhu is an academic researcher from University of Electronic Science and Technology of China. The author has contributed to research in topics: Computer science & Rank (linear algebra). The author has an hindex of 38, co-authored 280 publications receiving 5437 citations. Previous affiliations of Ce Zhu include National University of Singapore & Nanyang Technological University.
Papers
More filters
Journal ArticleDOI
Hexagon-based search pattern for fast block motion estimation
Ce Zhu,Xiao Lin,Lap-Pui Chau +2 more
TL;DR: Analysis shows that a speed improvement rate of the hexagon-based search (HEXBS) algorithm over the diamond search (DS) algorithm can be over 80% for locating some motion vectors in certain scenarios.
Proceedings ArticleDOI
Independently Recurrent Neural Network (IndRNN): Building A Longer and Deeper RNN
TL;DR: Independently Recurrent Neural Network (IndRNN) as discussed by the authors is a new type of RNN, where neurons in the same layer are independent of each other and they are connected across layers.
Proceedings ArticleDOI
Distribution-Aware Coordinate Representation for Human Pose Estimation
TL;DR: DARK as mentioned in this paper proposes a distribution-aware coordinate representation of keypoints (DARK) method, which improves the standard coordinate encoding process by generating unbiased/accurate heatmaps.
Posted Content
Independently Recurrent Neural Network (IndRNN): Building A Longer and Deeper RNN
TL;DR: Independently Recurrent Neural Network (IndRNN) as mentioned in this paper is a new type of RNN, where neurons in the same layer are independent of each other and they are connected across layers.
Journal ArticleDOI
Fast crowd density estimation with convolutional neural networks
TL;DR: This work proposes to estimate crowd density by an optimized convolutional neural network (ConvNet) first introduced for crowd density estimation, and introduces a cascade of two ConvNet classifier which improves both of the accuracy and speed.