Yan Xu

PhD Candidate
Electrical and Computer Engineering
Carnegie Mellon University
E-mail: yxu2@andrew.cmu.edu

About Me

My name is Yan Xu. Friends also call me Ryan. I am a Ph.D. student of CMU ECE Department advised by Prof. Kris Kitani. My research focus is camera pose estimation and 3D human pose estimation. Particularly, I am interested in combining the feature representation ability of neural networks with the precision of geometric optimization to develop pose estimation algorithms independent of dataset-specific training and can generalize to the in-the-wild setting. Application-wise, I feel excited about Auto-Driving and AR/VR/MR. So, thanks for visiting my page!

News

Research

Visio-Temporal Attention for Multi-Camera Multi-Target Association

Li, Yu-Jhe, Xinshuo Weng, Yan Xu, Kris Kitani.

ICCV 2021

[Paper pending][Code]
Wide-Baseline Multi-Camera Calibration using Person Re-Identification

Yan Xu Yu-Jhe Li, Xinshuo Weng, Kris Kitani.

CVPR 2021

[Paper][Code]
Few-Shot Learning with Intra-Class Knowledge Transfer

Yan Xu* Vivek Roy*, Yu-Xiong Wang, Kris Kitani, Ruslan Salakhutdinov, Martial Hebert.

Arxiv preprint 2020

[Paper][Code]
Estimating 3D Camera Pose from 2D Pedestrian Trajectories

Yan Xu, Vivek Roy and Kris Kitani

WACV 2020

[Paper][Code][Video]
Human-Interactive Subgoal Supervision for Efficient Inverse Reinforcement Learnin

Xinlei Pan, Eshed Ohn-Bar, Nicholas Rhinehart, Yan Xu, Yilin Shen, Kris M. Kitani

AAMAS 2018

[Paper]
Optimizing Quantiles in Preference-based Markov Decision Processes

Hugo Gilbert, Paul Weng and Yan Xu

AAAI 2017

[Paper]

Experience

Teaching

Fall 2021: 16720 - Computer Vision Capstone, given by Prof. Kris Kitani

Fall 2019: 18793 - Image and Video Processing, given by Prof. Aswin Sankaranarayanani

Fall 2018: 16622 - Master of Science in Computer Vision Capstone, given by Prof. Kris Kitani

Fall 2016: 18660 - Neumerical Methods for Engineering Designing and Optimization, given by Prof. Xin Li