Jia Xu, Ph.D.

Research Scientist
Intel Visual Computing Lab

Email: jiaxu [at] cs.wisc.edu
Picture of Jia Xu

Biography

I am a research scientist in the newly founded Intel Visual Computing Lab. I completed my Ph.D. in Computer Sciences at the University of Wisconsin-Madison, with my thesis committee of Prof. Vikas Singh (advisor), Prof. Chuck Dyer, Prof. Jerry Zhu, Prof. Jude Shavlik , and Prof. Mark Craven. I was a visiting student in University of Toronto during summer 2014, and in Toyota Technological Institute at Chicago during Summer 2013, both working with Prof. Raquel Urtasun. In the first year (2010-2011) of my graduate study, I worked as a research assistant/intern at Epic. Before graduate school, I obtained my B.S. degree from the Department of Computer Science and Technology at Nanjing University, China in June 2010.

Full Curriculum Vitae.

Research Interests

My major interests are computer vision, machine learning, and optimization. In particular, I am interested in systematically building reliable visual perception algorithms (e.g., semantic segmentation, dense video labeling) by mathematically modeling the physical properties of visual data (e.g., geometry, structure, context), and learning from the massive unlabeled or weakly labeled data available on the Internet. The ultimate goal of my research is to enable computers to perceive and reason at human level.

What's New

  • Aug. 2015: I joined Intel Labs as a research scientist in the newly founded Intel Visual Computing Lab.
  • July. 2015: I defended my thesis on Visual Parsing with Weak Supervision.
  • Apr. 2015: Our paper on Manifold-valued Dirichlet Processes is accepted to ICML 2015.
  • Mar. 2015: Two papers accepted to CVPR 2015.
  • Jan. 2015: I had a wonderful visit at CMU Robotics Institute and gave a talk in the VASC Seminar.
  • Jan. 2015: I will be TAing CS 766 Computer Vision for Spring 2015.
  • Sep. 2014: Our paper on Joint Visual and Textual Mining on Social Media is accepted by the PhD Forum in ICDM 2014.
  • Jun. 2014: Our paper on Large Scale Spectral Clustering is accepted by ECCV 2014 .
  • May. 2014: Our proposal for my ongoing research on Interactive Segmentation is funded by Adobe. Thank you, Adobe!
  • Mar. 2014: Our paper on Weakly Supervised Semantic Segmentation is accepted by CVPR 2014 (With Alex Schwing and Raquel Urtasun).
  • Feb. 2014: Our proposal for my ongoing research on GOSUS is accepted by NVIDIA Corporation (with a gift of one Tesla K40 GPU). This support is gratefully acknowledged!
  • Jan. 2014: I will be TAing CS 766 Computer Vision for Spring 2014.
  • Nov. 2013: Source Code for our Grassmannian Online Subspace Updates with Structured-sparsity project is released.
  • Oct. 2013: Source Code for our Interactive Segmentation and Contour Completion project is released.
  • Aug. 2013: Our paper on Grassmannian Online Subspace Updates with Structured-sparsity is accepted by ICCV 2013.
  • Jun. 2013: Dataset for our Interactive Segmentation and Contour Completion project is released.
  • May. 2013: Our poster on Interactive Segmentation is accepted by WARF Discovery Challenge Research Symposium, 2013.
  • Feb. 2013: Our paper on Interactive Segmentation and Contour Completion is accepted by CVPR 2013.
  • Jan. 2013: I am organizing the Computer Vision Reading Group for Spring 2013.
  • Jan. 2013: I will be the TA for Methods in Medical Image Analysis for Spring 2013.
  • Nov. 2012: Our cosegmentation research is featured on UWSMPH News.
  • Research Projects

    Gaze-enabled Egocentric Video Summarization


  • Gaze-enabled Egocentric Video Summarization via Constrained Submodular Maximization
  • Weakly Supervised Semantic Segmentation


  • Learning to Segment Under Various Forms of Weak Supervision
  • Tell Me What You See and I will Show You Where It Is
  • Structured Sparsity for Video Segmentation and Spectral Clustering


  • Spectral Clustering with a Convex Regularizer on Millions of Images.
  • GOSUS: Grassmannian Online Subspace Updates with Structured-sparsity
  • Interactive Segmentation and Contour Completion


  • Incorporating Topological Constraints within Interactive Segmentation and Contour Completion via Discrete Calculus
  • Cosegmentation


  • Analyzing the Subspace Structure of Related Images: Concurrent Segmentation of Image Sets
  • Random Walks based Multi-Image Cosegmentation: : Quasiconvexity Results and GPU-based Solutions
  •  

    Computer Sciences | UW Home
    Locations of visitors to this page
    eXTReMe Tracker