headshot_YCShih.jpeg

YiChang Shih
施易昌

Senior Staff Software Engineer, TLM

Google

yichangshih at csail dot mit dot edu

Curriculum Vitae            | Google Scholar | LinkedIn

 

I currently lead a small team at Google Pixel Camera Algorithm team to improve mobile camera imaging quality through Computational Photography, Computer Vision, and On-device Machine Learning techniques. We develop end-to-end solutions across engineering and research boundary.

 

** We are hiring!** If you are interested in computational photography, please send me an email. New grads are welcome.

Short Bio

I received my Ph.D. from MIT CSAIL in 2015, advised by Frédo Durand and Bill Freeman on Computational Photography, Computer Vision, and Computer Graphics. My PhD research was about  data-driven graphics. By leveraging repetitive visual-information and high quality examplars, my thesis explores advanced image editing algorithms. I've worked on hardware devices for imaging and enhancement during my master in MIT. Before Google, I was a Research Scientist at LightCo, where we launched the very first (and super cool) L16 camera (Verge).  

Luckily, I had great intern opportunities to work with Jon Barron and Marc Levoy (Google[x], 2014), Sylvain Paris and Connelly Barnes (Adobe Research , 2013), Vivek Kwatra and Troy Chinen (Google Research , 2012) Neel Joshi and Brian Guenter. (Microsoft Research, 2011). I am always happy to work with talented interns. Please contact me if you are interested in doing an internship with me. I have been lucky to work with the following interns so far:

  • Kai-En Lin: from UCSD, on Object NeRF (2021)

  • Chen Gao: from Virginia Tech, on Portrait NeRF (2020)

  • Zixun Yu: from Purdue, on portrait image enhancement (2019)

  • Wei-Shang Lai: from UC Merced, on wide-angle portrait distortion correction (2018)

Publications

Face Deblurring using Dual Camera Fusion on Mobile Phones

[webpage][pdf]

Wei-Sheng Lai, YiChang Shih, Lun-Cheng Chu, Xiaotong Wu, Sung-Fang Tsai, Michael Krainin, Deqing Sun, and Chia-Kai Liang

ACM Transactions on Graphics (SIGGRAPH) 2022

tip_video_face_correction.gif
IMG_20200206_142609.gif
IMG_20200205_140657.gif

Correcting Face Distortion in Wide-Angle Videos

[webpage] [pdf] [video]

 

Wei-Sheng Lai, YiChang Shih, Chia-Kai Liang, and Ming-Hsuan Yang

IEEE TIP 2021

Portrait Neural Radiance Fields from a Single Image

[webpage][pdf]

Chen Gao, YiChang Shih, Wei-Sheng Lai, Chia-Kai Liang, and Jia-Bin Huang

arXiv:2012.05903, 2021

representative.jpeg

Distortion-Free Wide-Angle Portraits on Camera Phones 

[project_web] [pdf]

YiChang Shih, Wei-Sheng Lai, Chia-Kai Liang

ACM Transactions on Graphics (SIGGRAPH) 2019

transform.jpeg

Transform Recipes for Efficient Cloud Photo Enhancement

[project_web] [pdf]

Michaël Gharbi, YiChang Shih, Gaurav Chaurasia, Jonathan Ragan-Kelley, Sylvain Paris, and Frédo Durand

ACM Transactions on Graphics (SIGGRAPH Asia) 2015

mit_logo.jpeg

PhD Thesis: Data-Driven Photographic Style Using Local Transfer

[thesis pdf][defense presentation (pdf)][pptx (171MB)]

YiChang Shih

March 2015, MIT CSAIL, Cambridge, MA

BarronCVPR2015_anim.gif

Fast Bilateral-Space Stereo for Synthetic Defocus 

[pdf] [supp]

Jon Barron, Andrew Adams, YiChang Shih, and Carlos Hernandez

IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2015 (Oral)

reflection_logo.jpeg
portrait_icon.jpeg

Reflection Removal using Ghosting Cues 

[pdf][supp][simulator][poster][code]

YiChang Shih, Dilip Krishnan, Frédo Durand, William T. Freeman

IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2015

[The Economist]

Style Transfer for Headshot Portraits

[project_web][pdf]

 

YiChang Shih, Sylvain Paris, Connelly Barnes, William T. Freeman, Frédo Durand

ACM Transactions on Graphics (SIGGRAPH) 2014

TArrow.png
face2.jpeg
time_lapse_thumb.jpeg
pinholes.jpeg
speckle.png
lightfield.jpeg

Seeing the Arrow of Time

[project web] [pdf]

L. Pickup, Z. Pan, D. Wei, Y. Shih, W. Freeman, A. Zisserman and B. Schölkopf

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

Joint Noise Level Estimation from Personal Photo Collections

[project web][pdf]

 

YiChang Shih, Vivek Kwatra, Troy Chinen, Hui Fang, Sergey Ioffe

IEEE International Conference on Computer Vision (ICCV) 2013

Data-driven Hallucination of Different Times of Day from a Single Outdoor Photo

[project web][pdf]

YiChang Shih, Sylvain Paris, Frédo Durand, William T. Freeman

ACM Transactions on Graphics (SIGGRAPH Asia) 2013

Image Enhancement using Calibrated Lens Simulations

[project web] [pdf]

 

YiChang Shih, Brian Guenter, Neel Joshi

European Conference on Computer Vision (ECCV) 2012

Laser Speckle Photography for Surface Tampering Detection

[project web] [pdf]

YiChang Shih, Abe Davis, Samuel W. Hasinoff, Frédo Durand, William T. Freeman

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

YiChang Shih, Master thesis, 2012 [pdf]

Light Field Analysis for Modeling Image Formation 

[pdf]

Chia-Kai Liang, YiChang Shih, Homer H. Chen

IEEE Trans. Image Processing. Vol. 20, No.2 pp. 446-460, 2011.

Course Projects

comfoto.png

Bars Removal by Multiview Images 

[pdf][ppt]

Course: 6.865 Computational Photography (2011)

with HsinJung Yang

sanskrit.jpeg

Handwritten Sanskrit Recognition using a Multi-class SVM with K-NN Guidance 

[pdf]

Course: 6.867 Machine Learning (2011)

with Donglai Wei

Teaching Assistant

icon.jpeg

6.815/6.865 - Digital & Computational Photo (2013)

 

[course_web]

Lecturer: Frédo Durand

Vision Transformer for NeRF-Based View Synthesis from a Single Input Image

[webpage][pdf]

Kai-En Lin, Yen-Chen Lin, Wei-Sheng Lai, Tsung-Yi Lin, YiChang Shih, and Ravi Ramamoorthi

arXiv:2207.05736, 2022