Home Education Publications Honors CV

Fei Xia

feixia at stanford.edu [Github] [Google Scholar]

About me

I am a fifth-year PhD candidate at the Department of Electrical Engineering, Stanford University. I am co-advised by Silvio Savarese in SVL and Leo Guibas. I am supported by Stanford Graduate Fellowship and Qualcomm Innovation Fellowship. During my PhD, I have done research internships with Dieter Fox at Nvidia, and Alexander Toshev and Brian Ichter at Google. I obtained my bachelor's degree from Tsinghua University in 2016.

Research Interest

My mission is to build intelligent embodied agents that can interact with complex and unstructured real-world environments, with applications to home robotics. I approach this problem from 3 aspects: 1) Large scale and transferrable simulation for Robotics. 2) Learning algorithms for long-horizon tasks. 3) Combining geometric and semantic representation for environments.

Outreach

On weekends, depending on availability, I voluntarily host office hours for students (especially underrepresented groups and junior students) who want to get into the field of and develop a career on Machine Learning, Computer Vision, and Robotics. Each slot is 20-minute long. If you want to get advice from me, please fill out this questionnaire.
News:

  • 2021.5 I defended my PhD Thesis titled "Large Scale Simulation for Embodied Perception and Robot Learning".

  • 2021.3 I will join Robotics at Google as a Research Scientist in the Fall.

  • 2021.3 ReLMoGen accepted to ICRA2021.

  • 2020.12 The wait is over! iGibson v1.0 was released! It comes with many new features and fully interactive environments, checkout the website for more details.

  • 2020.4 iGibson was released! It is a large scale interactive environment for robot learning.

  • 2020.3 I am co-hosting CVPR Challenge "Sim2Real Challenge with Gibson". It is the first Sim2Real challenge in CVPR.

  • 2019.7 AdaFDR was accepted to Nature Communications.

  • 2019.5 Will Shen and I won Qualcomm Innovation Fellowship as a team. Thank you Qualcomm!

  • 2019.5 AdaFDR won best paper award at RECOMB2019.

  • 2018.4 We will host a demo for Gibson Env at CVPR'18. Come and check it out!

  • 2018.2 One paper accepted to CVPR'18 (spotlight).

Education

Aug. 2016 - Present, Department of Electrical Engineering, Stanford University,

PhD Candidate.

Aug. 2012 - Jul. 2016, Department of Automation Tsinghua University,

Bachelor of Engineering.

Aug. 2014 - Dec. 2014, Department of Electrical and Computer Engineering, Georgia Institute of Technology,

Exchange Student.

July. 2015 - Sept. 2015, Department of Electrical Engineering, Stanford University,

Visiting Researcher.

Selected Publications

[full list]
  • Fei Xia*, Chengshu Li*, Roberto Martín-Martín, Or Litany, Alexander Toshev, Silvio Savarese (2020). ReLMoGen: Leveraging Motion Generation in Reinforcement Learning for Mobile Manipulation. ICRA2021. [pdf][project]
  • Bokui Shen*, Fei Xia*, Chengshu Li*, Roberto Martín-Martín*, Linxi Fan, Guanzhi Wang, Shyamal Buch, Claudia D'Arpino, Sanjana Srivastava, Lyne P. Tchapmi, Micael E. Tchapmi, Kent Vainio, Li Fei-Fei, Silvio Savarese. iGibson, a Simulation Environment for Interactive Tasks in Large Realistic Scenes. (under review) [pdf][project]
  • Fei Xia, William B Shen, Chengshu Li, Priya Kasimbeg, Micael Edmond Tchapmi, Alexander Toshev, Roberto Martín-Martín, Silvio Savarese. Interactive Gibson Benchmark: A Benchmark for Interactive Navigation in Cluttered Environments. ICRA20 + RAL. [pdf][project]
  • Noriaki Hirose, Fei Xia, Roberto Martín-Martín, Amir Sadeghian, Silvio Savarese (2019). Deep Visual MPC-Policy Learning for Navigation. IEEE Robotics and Automation Letters . [pdf][project]
  • Noriaki Hirose,Amir Sadeghian, Fei Xia, Roberto Martín-Martín, Silvio Savarese. (2019). VUNet: Dynamic Scene View Synthesis for Traversability Estimation using an RGB Camera. IEEE Robotics and Automation Letters. [pdf][project]

  • Fei Xia*, Amir R. Zamir*, Zhiyang He*, Alexander Sax, Jitendra Malik, Silvio Savarese. Gibson Env: Real-World Perception for Embodied Agents. CVPR 2018 (spotlight, Nvidia Pioneer Research Award). [pdf] [code] [project]

  • Fei Xia*, Martin Zhang*, James Zou, David Tse. NeuralFDR: learning decision threshold from hypothesis features. NIPS 2017. [pdf] [code]

Previous Computational Biology papers

  • Martin J. Zhang, Fei Xia, James Zou, "AdaFDR: a Fast, Powerful and Covariate-Adaptive Approach to Multiple Hypothesis Testing", Preliminary version selected as Best Paper Award of RECOMB 2019, also in Nature Communications, 2019. [pdf] [software] [code to reproduce the paper]

  • Qiao Liu, Fei Xia, Qijin Yin, Rui Jiang. Chromatin accessibility prediction via a hybrid deep convolutional neural network Bioinformatics. [pdf] [code]

  • Govinda M Kamath*, Ilan Shomorony*, Fei Xia*, Thomas A Courtade, David Tse. HINGE: long-read assembly achieves optimal repeat resolution. [pdf] [code] Genome Research Vol 27 2017.

  • Ilan Shomorony, Govinda M Kamath, Fei Xia, Thomas A Courtade, David Tse. Partial DNA assembly: a rate-distortion perspective. [pdf] ISIT 2016.

(* Equally contributed to the project.)

Honors and Awards

  • 2019 Qualcomm Innovation Fellowship
  • 2019 RECOMB Best Paper Award
  • 2018 Nvidia Pioneer Research Award at CVPR
  • 2016 Stanford Graduate Fellowship (Michael J. Flynn Fellow), Stanford University
  • 2015 Chang Jiong Scholarship (Highest honor in Dept. of Automation, 1/560)
  • 2014 Fang Chongzhi Scholarship (Highest honor in Dept. of Automation, 1/560)
  • 2014 China Scholarship Council Excellent Undergraduate Fellowship

Teaching Experiences

Press Coverage