Stella Yu

1947 Center St Suite 600 Berkeley, CA 94704
Lab Page
AFFILIATIONS Professor of Computer Science
Director, Vision Group at the International Computer Science Institute
Senior Fellow, Berkeley Institute for Data Science

Research Interests

My approach to computer vision is about establishing parts in relation, organizing them into a whole, and singling out the aspect of interest from the whole. In modeling visual organization, I have developed new tools in spectral graph theory and machine learning to effectively uncover the relational structure of large-scale data. I have also pursued efficient representation and computation by learning from artists and exploring the connections between Art and Vision.

Selected Publications

Image Quality Assessment by Comparing CNN Features Between Images. Seyed Ali Amirshahi and Marius Pedersen and Stella X. Yu. Journal of Imaging Science and Technology, 2016 Abstract | Paper

The Perception of Symmetry in the Moving Image — Multi-Level Computational Analysis of Cinematographic Scene Structure and Its Visual Reception. Jakob Suchan and Mehul Bhatt and Stella X. Yu.  ACM Symposium on Applied Perception, Anaheim, California, 22-23 July 2016  Abstract | Paper

Fine-to-Coarse Knowledge Transfer for Low-Res Image Classification.  Xingchao Peng and Judy Hoffman and Stella X. Yu and Kate Saenko.  International Conference on Image Processing, Phoenix, Arizona, 25-28 September 2016  Abstract | Paper | arXiv

Affinity CNN: Learning Pixel-Centric Pairwise Relations for Figure/Ground Embedding. Michael Maire and Takuya Narihira and Stella X. Yu.  IEEE Conference on Computer Vision and Pattern Recognition, Las Vegas, Nevada, 27-30 June 2016 Abstract | Paper | Slides | arXiv

Convolutional Random Walk Networks for Semantic Image Segmentation. Gedas Bertasius and Lorenzo Torresani and Stella X. Yu and Jianbo Shiar. Xiv:1605.07681, online, 24 May 2016  Abstract | arXiv

A Deep Neural Net Trained for Person Categorization Develops Both Detailed Local Features and Broad Contexual Specificities. Stella X. Yu and Karl Zipser. Vision Sciences Society Annual Meeting, St. Pete Beach, Florida, 13-18 May 2016  Abstract | Poster

First Person Action-Object Detection with EgoNet. Gedas Bertasius and Hyun Soo Park and Stella X. Yu and Jianbo Shiar. Xiv:1603.04908, online, 15 March 2016 Abstract | arXiv

Constrained Structured Regression with Convolutional Neural Networks. Deepak Pathak and Philipp Kraehenbuehl and Stella X. Yu and Trevor Darrellar. Xiv:1511.07497, online, 23 November 2015  Abstract | arXiv

Mapping Images to Sentiment Adjective Noun Pairs with Factorized Neural Nets. Takuya Narihira and Damian Borth and Stella X. Yu and Karl Ni and Trevor Darrellar. Xiv 1511.06838, online, 21 November 2015  Abstract | arXiv

Direct Intrinsics: Learning Albedo-Shading Decomposition by Convolutional Regression. Takuya Narihira and Michael Maire and Stella X. Yu. International Conference on Computer Vision, Santiago, Chile, 13-16 December 2015  Abstract | Paper | Poster | Code

Learning Lightness from Human Judgement on Relative Reflectance. Takuya Narihira and Michael Maire and Stella X. Yu. IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, 8-10 June 2015 Abstract | Paper | Poster

FlowWeb: Joint Image Set Alignment by Weaving Consistent, Pixel-wise Correspondences. Tinghui Zhou and Yong Jae Lee and Stella X. Yu and Alexei A. Efros. IEEE Conference on Computer Vision and Pattern Recognition, Boston, MA, 8-10 June 2015 Abstract | Paper | Slides | Dataset | Code

See more at: