Jiahui Yu, PhD

Research Scientist,
Google Brain

Google Scholar  /  GitHub
LinkedIn  /  Twitter  /  Facebook


I am a research scientist at Google Brain. I received my PhD at University of Illinois at Urbana-Champaign in 2020, advised by Professor Thomas Huang. Previously I received a Bachelor with distinction at School of the Gifted Young in Computer Science, University of Science and Technology of China in 2016. I have interned at Microsoft Research Asia, Face++/Megvii, Adobe Research, Snap Research, Jump Trading, Baidu Research, Nvidia Research, and Google Brain.

My research interest lies in visual perception, generative models, sequences, and high performance computing.


Publications

(Google Scholar Profile)

BigNAS: Scaling Up Neural Architecture Search with Big Single-Stage Models
Jiahui Yu, Pengchong Jin, Hanxiao Liu, Gabriel Bender, Pieter-Jan Kindermans, Mingxing Tan, Thomas Huang, Xiaodan Song, Ruoming Pang, Quoc Le.
ArXiv

FSNet: Compression of Deep Convolutional Neural Networks by Filter Summary
Yingzhen Yang, Jiahui Yu, Nebojsa Jojic, Jun Huan, Thomas Huang.
ICLR 2020 / OpenReview

Scale-wise Convolution for Image Restoration
Yuchen Fan, Jiahui Yu, Ding Liu, Thomas Huang.
AAAI 2020 / Code

AutoSlim: Towards One-Shot Architecture Search for Channel Numbers
Jiahui Yu and Thomas Huang.
ArXiv / Code

Universally Slimmable Networks and Improved Training Techniques
Jiahui Yu and Thomas Huang.
ICCV 2019 / Code

Free-Form Image Inpainting with Gated Convolution (DeepFill v2)
Jiahui Yu, Zhe Lin, Jimei Yang, Xiaohui Shen, Xin Lu, Thomas Huang.
ICCV 2019 (Oral Presentation) / Code / Live Demo!

Fast Proximal Gradient Descent for Non-convex Optimization
Yingzhen Yang and Jiahui Yu.
UAI 2019 / Code

Foreground-aware Image Inpainting
Wei Xiong, Jiahui Yu, Zhe Lin, Jimei Yang, Xin Lu, Connelly Barnes, Jiebo Luo.
CVPR 2019

Slimmable Neural Networks
Jiahui Yu, Linjie Yang, Ning Xu, Jianchao Yang, Thomas Huang.
ICLR 2019 / Code / OpenReview
Top-10 Rated Papers on OpenReview, ICLR 2019.

Wide Activation for Efficient and Accurate Image Super-Resolution
Jiahui Yu, Yuchen Fan, Jianchao Yang, Ning Xu, Zhaowen Wang, Xinchao Wang, Thomas Huang.
BMVC 2019 (challenge report) / Code
Won 1st in NTIRE Challenge on Single Image Super-Resolution, CVPR 2018.

Improving Object Detection from Scratch via Gated Feature Reuse
Zhiqiang Shen, Honghui Shi, Jiahui Yu, Hai Phan, Rogerio Feris, Liangliang Cao, Ding Liu, Xinchao Wang, Thomas Huang, Marios Savvides.
BMVC 2019 / Code

Generative Image Inpainting with Contextual Attention (DeepFill v1)
Jiahui Yu, Zhe Lin, Jimei Yang, Xiaohui Shen, Xin Lu, Thomas Huang.
CVPR 2018 / Code / Live Demo! (10,000+ trials in 72 hours after release)

Neighborhood Regularized l1-Graph
Yingzhen Yang, Jiashi Feng, Jiahui Yu, Jianchao Yang, Pushmeet Kohli, Thomas Huang.
UAI 2017

Support Regularized Sparse Coding and Its Fast Encoder
Yingzhen Yang, Jiahui Yu, Pushmeet Kohli, Jianchao Yang, Thomas Huang.
ICLR 2017

UnitBox: An Advanced Object Detection Network
Jiahui Yu, Yuning Jiang, Zhangyang Wang, Zhimin Cao, Thomas Huang.
ACMMM 2016
Adapted to official TensorFlow Object Detection API.


Awards

  • Baidu Scholarship
  • Thomas and Margaret Huang Research Award
  • Microsoft-IEEE Young Fellowship

Service

  • Journal Reviewer: TPAMI, IJCV, TIP, TMM, TMI, TVCG, TVCJ, TCSVT, TSC, TGRS, IMAGE, JSTSP, JVCI, NEUCOM, PR, SPL, MULT, TNNLS, ENG.
  • Conference Reviewer: CVPR, ICCV, ECCV, NeurIPS, ICLR, SIGGRAPH, ACM Multimedia, AAAI, IJCAI, ICME, PRCV.