J
Junyu Dong
Researcher at Ocean University of China
Publications - 484
Citations - 6570
Junyu Dong is an academic researcher from Ocean University of China. The author has contributed to research in topics: Computer science & Feature extraction. The author has an hindex of 30, co-authored 399 publications receiving 3553 citations. Previous affiliations of Junyu Dong include Qingdao University.
Papers
More filters
Proceedings ArticleDOI
Multiplex Heterogeneous Graph Convolutional Network
TL;DR: The proposed Multiplex Heterogeneous Graph Convolutional Network (MHGCN) can automatically learn the useful heterogeneous meta-path interactions of different lengths in multiplex heterogeneous networks through multi-layer convolution aggregation and effectively integrates both multi-relation structural signals and attribute semantics into the learned node embeddings with both unsupervised and semi-supervised learning paradigms.
Journal ArticleDOI
Self-attention neural architecture search for semantic image segmentation
TL;DR: Li et al. as mentioned in this paper investigated the self-attention along all the possible dimensions (H,W,C,HW,HC,CW,HWC) and applied the neural architecture search (NAS) technique to achieve optimal aggregation.
Posted Content
A PCA-Based Convolutional Network.
TL;DR: The results show that PCN performs competitive with or even better than state-of-the-art deep learning models and since there is no back propagation for supervised finetuning, PCN is much more efficient than existing deep networks.
Journal ArticleDOI
Texture Classification Using Pair-Wise Difference Pooling-Based Bilinear Convolutional Neural Networks
TL;DR: Since the dimensionality of the BCNN feature vectors is very high, a new yet simple Block-wise PCA (BPCA) method is proposed in order to derive more compact feature vectors.
Proceedings ArticleDOI
An Approach for Quantitative Evaluation of the Degree of Facial Paralysis Based on Salient Point Detection
TL;DR: A new approach for quantitatively estimating the degree of facial paralysis is described, which first determine key points based on salient point detection and then the differences between the two facial sides are calculated.