Jun-Yan Zhu

Ph.D. candidate

Berkeley AI Research (BAIR) Lab

Department of EECS

University of California, Berkeley

Email:

junyanz at eecs dot berkeley dot edu

Office:

Sutardja Dai Hall 7th floor
University of California, Berkeley
Berkeley, CA 94704

 

[CV] [GitHub] [Google Scholar]

 

I am a Ph.D. student at Berkeley AI Research Lab. Before coming here, I was a Ph.D. student in CS Department at CMU. I do research on computer vision, graphics, and machine learning with Professor Alexei A. Efros. My research goal is to build machines capable of recreating our visual world. I am currently supported by a Facebook Fellowship.


I received my B.E in Computer Science from Tsinghua University in 2012, where I worked with Professor Zhuowen Tu and Dr. Eric Chang at Microsoft Research Asia. I was also a member of Tsinghua's Graphics Group led by Professor Shi-Min Hu.

Cat Papers

If you like cats, and love reading cool vision, learning and graphics papers, check out [GitHub] [Webpage].

Talks

Visual Manipulation and Synthesis on the Natural Image Manifold

Facebook, Berkeley BAIR, Tsinghua, MSR, Fudan Univ, ICML 16' workshop "Visualization for Deep Learning" (2016)

Mirror Mirror: Crowdsourcing Better Portraits

ACM SIGGRAPH Asia 2014 (Dec 2014)

What Makes Big Visual Data Hard?

ACM SIGGRAPH Asia 2014 invited course "Data-Driven Visual Computing" (Dec 2014)

AverageExplorer: Interactive Exploration and Alignment of Visual Data Collections

ACM SIGGRAPH 2014 (Aug 2014)

Discovering Objects and Harvesting Visual Concepts via Weakly Supervised Learning

Berkeley Visual Computing Lab Noon Talk (Mar 2014)


Software

Interactive Deep Colorization: python implementation for real-time user-guided colorization

Light Field Video: Light field video applications (e.g. video refocusing, focus tracking, changing aperture and view)

CycleGAN: Torch implementation for learning an image-to-image translation without input-output pairs

pix2pix: Torch implementation for learning a mapping from input images to output images

pytorch CycleGAN & pix2pix: PyTorch implementation for both unpaired and paired image-to-image translation

iGAN: a deep learning software that easily generates images with a few brushstrokes

RealismCNN: code for predicting and improving visual realism in composite images

MCILBoost: a boosting-based Multiple Instance Learning (MIL) package that includes MIL-Boost and MCIL-Boost

MirrorMirror: an expression training App that helps users mimic their own expressions

SelectGoodFace: a program for selecting attractive/serious portraits from a personal photo collection

FaceDemo: a simple 3D face alignment and warping demo


Publications

Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks

Jun-Yan Zhu*, Taesung Park*, Phillip Isola, and Alexei A. Efros

In Arxiv, 2017. (*indicates equal contributions)

 

[Project] [GitHub] [PyTorch]
[Paper] [BibTex]

 

Real-Time User-Guided Image Colorization with Learned Deep Priors

Richard Zhang*, Jun-Yan Zhu*, Phillip Isola, Xinyang Geng, Angela S. Lin, Tianhe Yu, and Alexei A. Efros

In ACM Transactions on Graphics (conditionally accepted to SIGGRAPH), 2017

(*indicates equal contributions)

[Project] [GitHub] [Youtube]
[Paper] [Talk] [BibTex]

Light Field Video Capture Using a Learning-Based Hybrid Imaging System

Ting-Chun Wang, Jun-Yan Zhu, Nima Khademi Kalantari, Alexei A. Efros, and Ravi Ramamoorthi

In ACM Transactions on Graphics (conditionally accepted to SIGGRAPH), 2017

[Project] [GitHub] [Youtube] [Training/Test code]
[Paper] [Video] [Data (18GB)] [BibTex]

Image-to-Image Translation with Conditional Adversarial Nets

Phillip Isola, Jun-Yan Zhu, Tinghui Zhou, and Alexei A. Efros

In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017

See neat uses of #pix2pix on Twitter.

 

[Project] [GitHub] [PyTorch]
[Paper] [BibTex] [Two Minute Papers]

 

Generative Visual Manipulation on the Natural Image Manifold

Jun-Yan Zhu, Philipp Krähenbühl, Eli Shechtman, and Alexei A. Efros

In European Conference on Computer Vision (ECCV), 2016

See article in California Magazine

 

[YouTube] [Project] [GitHub] [Paper]
[Slides] [Video] [BibTex] [Two Minutes Papers]

 

A 4D Light-Field Dataset and CNN Architectures for Material Recognition

Ting-Chun Wang, Jun-Yan Zhu, Ebi Hiroaki, Manmohan Chandraker, Alexei A. Efros, and Ravi Ramamoorthi

In European Conference on Computer Vision (ECCV), 2016

 

[Paper] [Data (thumbnail)] [Full Data (15.9G)]

[Supplement] [Poster] [BibTex]

 

Learning a Discriminative Model for the Perception of Realism in Composite Images

Jun-Yan Zhu, Philipp Krähenbühl, Eli Shechtman, and Alexei A. Efros

In IEEE International Conference on Computer Vision (ICCV), 2015

 

[Project] [Paper] [GitHub]

[Slides] [Poster] [BibTex]

 

Mirror Mirror: Crowdsourcing Better Portraits

Jun-Yan Zhu, Aseem Agarwala, Alexei A. Efros, Eli Shechtman, and Jue Wang

In ACM Transactions on Graphics (SIGGRAPH Asia), 2014

 

[Project (with Code) ] [Paper] [Data]

[Slides] [Supplement] [BibTex]

AverageExplorer: Interactive Exploration and Alignment of Visual Data Collections

Jun-Yan Zhu, Yong Jae Lee and Alexei A. Efros

In ACM Transactions on Graphics (SIGGRAPH), 2014

 

See article in The New Yorker

[Project] [YouTube] [Paper]

[Slides] [Supplement] [BibTex]

 

MILCut: A Sweeping Line Multiple Instance Learning Paradigm for Interactive Image Segmentation

Jiajun Wu*, Yibiao Zhao*, Jun-Yan Zhu, Siwei Luo and Zhuowen Tu

In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2014

(*indicates equal contributions)

 

[Project] [Paper] [Poster] [BibTex]

 

Reverse Image Segmentation: A High-Level Solution to a Low-Level Task

Jiajun Wu, Jun-Yan Zhu, and Zhuowen Tu

In British Machine Vision Conference (BMVC), 2014

 

[Paper] [BibTex]

 

Unsupervised Object Class Discovery via Saliency-Guided Multiple Class Learning

Jun-Yan Zhu, Jiajun Wu, Yan Xu, Eric Chang and Zhuowen Tu

In IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2015

(an expanded journal version of our CVPR 2012 paper)

 

[Project] [Paper]

[Supplement] [Poster] [BibTex]

 

Multiple Clustered Instance Learning for Histopathology Cancer Image Classification, Segmentation and Clustering

Yan Xu*, Jun-Yan Zhu*, Eric I-Chao Chang and Zhuowen Tu

In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2012

(*indicates equal contributions)

(See an expanded journal version at Medical Image Analysis (MIA), 2014

 

[Project] [GitHub] [Paper]
[BibTex] [Poster]

 

Motion-Aware Gradient Domain Video Composition

Tao Chen, Jun-Yan Zhu, Ariel Shamir and Shi-Min Hu

In IEEE Transactions on Image Processing (TIP), 2013

 

[Paper] [Youtube] [Video] [BibTex]

 

 

 

Awards


Facebook Fellowship (2015-2017)

Outstanding Undergraduate Thesis in Tsinghua University (2012)

Excellent Undergraduate Student in Tsinghua University (2012)

National Scholarship, by Ministry of Education of China (2009 and 2010)

Singapore Technologies Engineering China Scholarship (2010, 2011, and 2012)



Patents


US 20140270495. Multiple Cluster Instance Learning for Image Classification

US 20140140610. Unsupervised Object Class Discovery via Bottom Up Multiple Class Learning



MISC


Here is my cat Aquarius.



web page statistics from GoStats

Since August 2014