Jia Xu, Ph.D.

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


I am the Head of Huya AI and General Manager at Huya Live. I was a principal researcher and manager at Tencent AI Lab. Before returning to China, I was a senior research scientist in the Intel Visual Computing Lab, lead by the awesome Vladlen Koltun. I received 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 and in Toyota Technological Institute at Chicago, both working with Prof. Raquel Urtasun. Before graduate school, I obtained my B.S. degree from the Department of Computer Science and Technology at Nanjing University, China.

  • I am hiring motivated researchers/engineers/interns. If you work on computer vision, deep learning, speech recognition, or computer graphics, feel free to contact me.
  • What's New

  • Jun. 2019: Our SelFlow paper was selected in the CVPR Best Paper Finalist.
  • Mar. 2019: Our Self-Supervised Optical Flow Learning paper was selected for full oral presentation at CVPR 2019. This is the Winner entry of Sintel Optical Flow Benchmark since 11/2018.
  • Feb. 2019: Two papers were accepted to CVPR 2019. Congratulations to my interns Pengpeng and Jing, these are their very first CVPR papers!
  • Dec. 2018: Our DHER paper was accepted to ICLR 2019. Congratulations to team!
  • Nov. 2018: Our DDFlow paper was accepted to AAAI 2019 as oral presentation. Congratulations to my intern, Pengpeng! This is Pengpeng's very first paper published at a top tier conference.
  • Aug. 2018: Together with my intern Shiyu and collaborators at Tsinghua University, we won the Visual Doom AI Competition 2018.
  • Feb. 2018: Our Learning to See in the Dark paper was accepted to CVPR 2018.
  • July 2017: Our Fast Image Processing paper was accepted to ICCV 2017.
  • Feb. 2017: Our DCFlow paper was accepted to CVPR 2017. Winner on the Sintel Optical Flow Benchmark from 11/2016 to 11/2017.
  • Research Projects

    Optical Flow

  • SelFlow: Self-Supervised Learning of Optical Flow
  • DDFlow: Learning Optical Flow with Unlabeled Data Distillation
  • Accurate Optical Flow via Direct Cost Volume Processing
  • Fast Image Processing

  • Learning to See in the Dark
  • Fast Image Processing with Fully-Convolutional Networks
  • Semantic Segmentation

  • Learning to Segment Under Various Forms of Weak Supervision
  • Tell Me What You See and I will Show You Where It Is
  • Gaze-enabled Egocentric Video Summarization

  • Gaze-enabled Egocentric Video Summarization via Constrained Submodular Maximization
  • 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