J
Junchi Yu
Researcher at Chinese Academy of Sciences
Publications - 17
Citations - 73
Junchi Yu is an academic researcher from Chinese Academy of Sciences. The author has contributed to research in topics: Computer science & Medicine. The author has an hindex of 2, co-authored 3 publications receiving 23 citations.
Papers
More filters
Proceedings ArticleDOI
Pose-preserving Cross Spectral Face Hallucination
TL;DR: This work presents an approach to avert the data misalignment problem and faithfully preserve pose, expression and identity information during cross-spectral face hallucination and outperforms current state-of-the-art HFR methods at a high resolution.
Journal ArticleDOI
Structure-aware conditional variational auto-encoder for constrained molecule optimization
TL;DR: In this article , a structure-aware conditional variational auto-encoder (SCVAE) is proposed, which exploits the topology of molecules as structure condition and optimizes the molecular properties with constrained structural modification.
Journal ArticleDOI
A Survey of Trustworthy Graph Learning: Reliability, Explainability, and Privacy Protection
Bingzhe Wu,Jintang Li,Junchi Yu,Yatao An Bian,Hengtong Zhang,Chaochao Chen,Chengbin Hou,Qiang Chen,Tingyang Xu,Yu Rong,Xiaojing Zheng,Junzhou Huang,Ran He,Baoyuan Wu,Guangyu Sun,Peng Cui,Zibin Zheng,Zhe Liu,Peilin Zhao +18 more
TL;DR: A comprehensive review of recent leading approaches in the TwGL from three dimensions, namely, reliability, explainability, and privacy protection, which gives a general categorization for existing work and review typical work for each category.
Proceedings ArticleDOI
Semantic-Aware Makeup Cleanser
TL;DR: The experimental results verify that SAMC not only produces appealing de-makeup outputs at a resolution of 256 × 256, but also facilitates makeup-invariant face verification through image generation.
Journal ArticleDOI
Finding Diverse and Predictable Subgraphs for Graph Domain Generalization
Junchi Yu,Jian Liang,Ran He +2 more
TL;DR: This paper proposes a new graph domain generalization framework, dubbed as DPS, by constructing multiple populations from the source domains, that is model-agnostic that can be incorporated with various GNN backbones.