Ph.D.
Robotics Institute
School of Computer Science
Carnegie Mellon University

Office: Smith Hall 210
Email: xinshuow@cs.cmu.edu

 total visits since 06/20/2017
Free Web Counters unique visits since 06/20/2017

I am a Ph.D. student (2018-) at the Robotics Institute of Carnegie Mellon University (CMU) supervised by Kris Kitani. I received my Masters (2016-17) at the CMU Robotics Institute, where I was working with Yaser Sheikh and Kris Kitani. Before entering my Ph.D. program at CMU, I worked at Oculus Research Pittsburgh (now Facebook Reality Lab) as a research engineer. My Bachelor's degree was received from the School of Electronic Information at Wuhan University in China.

Research Interests
Fields: Computer Vision, Machine Learning, Robotics, Multimedia
Topics: 3D Computer Vision, Autonomous Driving, Graph Neural Networks, Generative Modeling, Video Analysis

Resume | Google Scholar | GitHub | LinkedIn | Twitter | Facebook | ResearchGate | Semantic Scholar

News

  • 08/2020 - Three papers accepted at ECCV 2020 Workshops
  • 06/2020 - Two papers accepted at IROS 2020
  • 06/2020 - Keynote Talk at CVPR 2020 Workshop: Scalability in Autohomous Driving [Link]
  • 04/2020 - One paper accepted at TPAMI 2020
  • 03/2020 - One paper accepted at CVPR 2020
  • 08/2019 - One paper accepted at ICCV Workshops 2019
  • 06/2019 - We release the code for our 3D MOT paper here
  • See More
  • 06/2019 - One paper accepted at BMVC 2019
  • 06/2019 - One paper accepted at ACMMM 2019
  • 06/2019 - One paper accepted at IROS 2019
  • 01/2018 - One paper accepted at CVPR 2018
  • 10/2017 - One paper accepted at WACV 2018

  • Education

  • Carnegie Mellon University, Ph.D. in Robotics 2018 - Present
  • Carnegie Mellon University, M.S. in Computer Vision 2016 - 2018
  • Wuhan University, B.S. in Electrical Engineering 2012 - 2016

  • Industry Experience

  • Oculus Research Pittsburgh, Research Engineer 2018
  • Facebook, Research Intern 2017

  • Awards and Honors

  • Qualcomm Innovation Fellowship Finalist 2020
  • Microsoft Research Ada Lovelace Fellowship Nomination 2019
  • Google PhD Fellowship Nomination 2019

  • Teaching

  • Computer Vision (16-385), CMU 2019
  • Geometry-Based Methods in Computer Vision (16-822), CMU 2018

  • Professional Activity

    Journal Reviewer
  • TPAMI 2020
  • MTA 2019, 2020
  • TCSVT 2018

  • Conference Reviewer
  • CVPR 2018, 2020
  • ECCV 2020
  • ICCV 2019
  • ICLR 2021
  • ICML 2020
  • NeurIPS 2020
  • AAAI 2020, 2021
  • ICRA 2020
  • IROS 2020
  • WACV 2020, 2021
  • BMVC 2020
  • ACCV 2018, 2020
  • IV 2020

  • Workshop Reviewer
  • CVPR, AI City Challenge 2020

  • University Activity
  • MSCV Admission Committee, CMU 2019, 2020

  • Talks
  • CVPR, Keynote at Scalability in Autonomous Driving Workshop [Link] 2020
  • CMU, R-PAD (Robots Perceiving and Doing) Lab 2018
  • WACV, Oral 2018
  • Occlusion Reasoning for Multi-Object Tracking with Ghost Bounding Box Regression
    Jingjing Pan, Xinshuo Weng, Kris Kitani

    All-in-One Drive: A Large-Scale and Comprehensive Perception Dataset with High-Density Long-Range Point Cloud
    Xinshuo Weng, Yunze Man, Dazhi Cheng, Jinhyung Park, Matthew O'Toole, Kris Kitani
    arXiv, 2020
    PDF | Code | Demo | Website | Poster | Slides | BibTex
    The largest autonomous driving dataset with a super-set of synthetic sensors, complete annotations for all perception tasks and rare driving situations


    Unsupervised Sequence Forecasting of 100,000 Points for Unsupervised Trajectory Forecasting
    Xinshuo Weng, Jianren Wang, Sergey Levine, Kris Kitani, Nick Rhinehart
    arXiv:2003.08376, 2020
    PDF | Code | Demo | Website | Poster | Slides | BibTex
    By learning to forecast future LiDAR point clouds, we build a brand-new pipeline for trajectory forecasting without requiring trajectory labels

    Joint 3D Tracking and Forecasting with Graph Neural Network and Diversity Sampling
    Xinshuo Weng*, Ye Yuan*, Kris Kitani
    arXiv:2003.07847, 2020
    PDF | Code | Demo | Website | Poster | Slides | BibTex
    The first unified 3D MOT and trajectory forecasting method with object interaction modeling and diverse trajectory samples

    GNN3DMOT: Graph Neural Network for 3D Multi-Object Tracking with 2D-3D Multi-Feature Learning
    Xinshuo Weng, Yongxin Wang, Yunze Man, Kris Kitani
    IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020
    PDF | Code | Demo | Website | Poster | Slides | BibTex
    The first multi-object tracking method that leverages Graph Neural Network for object interaction modeling

    3D Multi-Object Tracking: A Baseline and New Evaluation Metrics
    Xinshuo Weng, Jianren Wang, David Held, Kris Kitani
    IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2020
    (Oral Presentation)
    PDF | Code | Demo | Website | Poster | Slides | BibTex
    A 3D multi-object tracker that achieves state-of-the-art performance with the fastest speed

    Monocular 3D Object Detection with Pseudo-LiDAR Point Cloud
    Xinshuo Weng, Kris Kitani
    IEEE International Conference on Computer Vision (ICCV) Workshops, 2019.
    PDF | Code | Demo | Website | Poster | Slides | BibTex
    By projecting the 2D image to a pseudo-LiDAR point cloud representation, our monocular 3D detection pipeline quadruples the performance over prior art

    Back to top