Keyu Li

Keyu Li (李柯宇)
Affiliation: Perception Group, Horizon Robotic
Address: Anting Town, Jiading District, Shanghai, China
Email:likeyuchn[at]gmail.com, likeyu[at]stu.xidian.edu.cn

About Me ([GitHub] [Google Scholar] [CV])

I am an Algorithm Engineer at Horizon Robotic.

Previously, I obtained the Master degree from Electronic and Information Engineering, Xidian University (XDU) in 2023. My academic supervisor is Prof. Nannan Wang. I received my B.E degree from Xidian University (XDU) in 2016.

My recent works are mainly on:

Autonomous Driving Perception
3D Object Detection & Tracking
Binary Neural Networks
Single Image Super-Resolution

News

2023/06: Our work UniAD wins CVPR 2023 Best Paper Award.
2023/06: I obtained Outstanding Master Thesis Award at Xidian University.
2023/03: One paper accepted to CVPR 2023.
2022/10: Two papers accepted to TNNLS.
2021/01: One paper accepted to AAAI 2021.
2020/07: One paper accepted to PRCV 2020, and selected as Oral Presentation.

Experience

2023/07 - Present: Algorithm Engineer at Horizon Robotic, responsible for perception algorithms in autonomous driving.
2021/06 - 2022/02: Research Intern at Shanghai AI Laboratory, led by Li Chen and Dr. Hongyang Li, research on end-to-end perception and decision model in autonomous driving.
2021/07 - 2022/05: Research Intern at Megvii Limited, research on lidar-based end-to-end 3D perception and vision-based 3D detection in autonomous driving.
2020/07 - 2021/06: Research collaboration with Huawei Noah's Ark Laboratory, research on binary neural networks and single image super-resolution.

Selected Works

(* Equal contribution, † Interns, # Corresponding authors)

Planning-oriented Autonomous Driving
Yihan Hu*, Jiazhi Yang*, Li Chen*#, Keyu Li*†, Chonghao Sima, Xizhou Zhu, Siqi Chai, Senyao Du, Tianwei Lin, Wenhai Wang, Lewei Lu, Xiaosong Jia, Qiang Liu, Jifeng Dai, Yu Qiao, Hongyang Li#;
CVPR, 2023 (Best Paper Award)
[Paper] [Code]

[Report] [BibTex]

First time to unify full-stack AD tasks!

Local Means Binary Networks for Image Super-Resolution
Keyu Li, Nannan Wang, Jingwei Xin, Xinrui Jiang, Jie Li, Xinbo Gao, Kai Han, Yunhe Wang
TNNLS, 2022
[Paper] [BibTex]

A binary quantization method for image super-resolution.

Feature Space based Loss for Face Photo-Sketch Synthesis
Keyu Li, Nannan Wang, Jie Li, Xinbo Gao
PRCV, 2020 (Oral Presentation)
[Paper] [BibTex]

Proposed new feature space based loss for face photo-sketch synthesis.

Training Binary Neural Network without Batch Normalization for Image Super-Resolution
Xinrui Jiang, Nannan Wang#, Jingwei Xin, Keyu Li, Xi Yang, Jie Li, Xinbo Gao
AAAI, 2021
[Paper] [BibTex]
Proposed a new binary training mechanism for binary image super-resolution.

Toward Pixel-Level Precision for Binary Super-Resolution With Mixed Binary Representation
Xinrui Jiang, Nannan Wang#, Jingwei Xin, Keyu Li, Xi Yang, Jie Li, Xinbo Gao
TNNLS, 2022
[Paper] [BibTex]

Mixed binary representation for binary image super-resolution.

Honors and Awards

2023/06: CVPR 2023 Best Paper Award
2023/06: Outstanding Master Thesis Award, Xidian University
2023/05: 2022-2023 Outstanding Graduate Student
2020/09: Outstanding Graduate Special Scholarship

Review Services

Program Committee Member/Conference Reviewer
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023
IEEE International Conference on Computer Vision (ICCV), 2023
CAAI International Conference on Artificial Intelligence (CICAI), 2021, 2022