Xingyu Chen (陈星宇)

I am passionate about bridging the physical and information worlds, and advancing next-generation AR/robotics with 3D spatial intelligence. I obtained my M.S. and B.S. from the Institute of Artificial Intelligence and Robotics at Xi’an Jiaotong University and Chongqing University.

In 2021 and 2022, I was a research intern at Tencent AI lab, working with Xuan Wang and Qi Zhang, hosted by Jue Wang.

My research topics include computer vision, graphics, and neural fields, specifically in inferring the physical world (shape, appearance, motion, etc.) and camera poses from images.

Email  |  Google Scholar  |  GitHub

headshot
News
Research

Equal Contribution * , Corresponding Author †

L2G-NeRF: Local-to-Global Registration for Bundle-Adjusting Neural Radiance Fields
Yue Chen*, Xingyu Chen*, Xuan Wang†, Qi Zhang, Yu Guo†, Ying Shan, Fei Wang
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023
project page / arXiv / paper / code / supplementary / video / poster

We combine local and global alignment via differentiable parameter estimation solvers to achieve robust bundle-adjusting Neural Radiance Fields.

UV Volumes for Real-time Rendering of Editable Free-view Human Performance
Yue Chen*, Xuan Wang*, Xingyu Chen, Qi Zhang, Xiaoyu Li, Yu Guo†, Jue Wang, Fei Wang
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023
project page / arXiv / paper / code / supplementary / video / poster

We separate high-frequency human appearance from 3D volume and encode them into a 2D texture, which enables real-time rendering and retexturing.

PROCA: Place Recognition under Occlusion and Changing Appearance via Disentangled Representations
Yue Chen, Xingyu Chen†, Yicen Li
IEEE International Conference on Robotics and Automation (ICRA), 2023
arXiv / code / video / poster

We decompose the image representation into place, appearance, and occlusion code and use the place code as a descriptor to retrieve images.

Sparse Semantic Map-Based Monocular Localization in Traffic Scenes Using Learned 2D-3D Point-Line Correspondences
Xingyu Chen, Jianru Xue†, Shanmin Pang
IEEE Robotics and Automation Letters (RA-L), 2022
arXiv / paper

Given a sparse semantic map (e.g., pole lines, traffic sign midpoints), we estimate camera poses by learning 2D-3D point-line correspondences.

Ha-NeRF😆: Hallucinated Neural Radiance Fields in the Wild
Xingyu Chen, Qi Zhang†, Xiaoyu Li, Yue Chen, Ying Feng, Xuan Wang, Jue Wang
IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022
project page / arXiv / paper / supplementary / code / video / poster

We recover NeRF from tourism images with variable appearance and occlusions, consistently rendering free-occlusion views with hallucinated appearances.

Using Detection, Tracking and Prediction in Visual SLAM to Achieve Real-time Semantic Mapping of Dynamic Scenarios
Xingyu Chen, Jianru Xue†, Jianwu Fang, Yuxin Pan, Nanning Zheng
IEEE Intelligent Vehicles Symposium (IV), 2020
arXiv / paper

Instead of detecting objects in all frames, we detect only keyframes and embed an efficient prediction mechanism in all incoming frames.

Navigation Command Matching for Vision-based Autonomous Driving
Yuxin Pan, Jianru Xue†, Pengfei Zhang, Wanli Ouyang, Jianwu Fang, Xingyu Chen
IEEE International Conference on Robotics and Automation (ICRA), 2020
ResearchGate / paper

We propose a navigation command matching model to discriminate actions generated from sub-optimal policies via smooth rewards.


Projects
Kuafu (Autonomous Vehicle)
IEEE Intelligent Transportation Systems Institute Lead Award
Second Place for Navigating in GPS-denied Environments, Intelligent Vehicle Future Challenge, 2019

Lidar Odometry, Mapping, and Localization.

Robotic Hand
Xingyu Chen, Zhili Liu, Yi Liu, Yang Li, Chenyang Liu
National Second Prize, The 11th Chinese College Students Computer Design Contest, 2018

Sensor fusion of IMU and BLE for localization.
Hand gesture recognition based on computer vision.
Robot hand controller based on potentiometer.

Robotic Arm
Xingyu Chen, Zhili Liu, Yang Li
National Second Prize, The 11th iCAN International Contest of innovAtioN, 2017

Hand gesture recognition based on computer vision.
Structure design based on four-bar linkage.


Invited Talks
光影幻象:神经辐射场中的时空流转
Xingyu Chen
Shenlan College online education, 2022

Introduction about Neural Radiance Fields (NeRF) for unconstrained photo collections.
Including NeRF, NeRF in the Wild, and Ha-NeRF

Academic Services
  • Conference Reviewer: CVPR, ICCV, NeurIPS, ICLR

template adapted from this awesome website
Last updated: Mar 2023