Jiefeng Li

I'm a Research Scientist in the Digital Human AI Research (DAIR) team at NVIDIA Research. I received my Ph.D. in the Department of Computer Science at Shanghai Jiao Tong University (SJTU) in 2024. I was a member of SJTU MVIG lab under the supervision of Prof. Cewu Lu. Prior to that, I received my Bachelor degree of Electronic Engineering in 2018 and Master degree of Computer Science in 2020 from Shanghai Jiao Tong University.

I'm interested in computer vision, machine learning, and generative AI. My current research mainly lies in reconstruction and animation of Digital Human.

Email / Google Scholar / Twitter / Github

News

July	2024	Two papers, COIN and DVT (Oral), accepted to ECCV 2024.
July	2024	Join NVIDIA Research as a Research Scientist.
May	2024	Defend my Ph.D. thesis, research on 3D Human Motion Estimation and Generation with Multi-Granularity Information.
June	2023	Start my internship at NVIDIA Research.
Feb	2023	Our paper NIKI on neural inverse kinematics is accepted to CVPR 2023.
Dec	2022	Tech Talk about AlphaPose at TechBeat.
Nov	2022	AlphaPose is accepted by TPAMI.
Oct	2022	I am invited to give a talk at IDEA.
Sep	2022	Jittor version of AlphaPose is released.
July	2022	Our paper D&D on learning dynamics properties of human motion is accepted to ECCV 2022 (Oral).
July	2022	Our paper SmoothNet and DLSA are accepted in ECCV 2022.
June	2022	Talk: Regression From the Perspective of MLE (Link).
Sep	2021	Our paper Sampling-Argmax is accepted in NeurIPS 2021.
Sep	2021	Invited Talk at Max-Planck Institute, Perceiving Systems Department.
July	2021	CVPR 2021特别企划: 基于混合逆运动学解法的3D人体形态估计 (Video).
July	2021	Our paper Residual Log-likelihood Estimation is accepted in ICCV 2021 (Oral).
July	2021	Our paper CPF is accepted in ICCV 2021.
Mar	2021	Our paper HybrIK is accepted in CVPR 2021.
May	2020	Our paper HMOR is accepted in ECCV 2020 (Spotlight).
Feb	2020	Our paper 2D-3D Joint HOI Learning is accepted in CVPR 2020.
July	2019	AlphaPose is included in GluonCV model zoo now.
Feb	2019	Our Paper CrowdPose is accepted in CVPR 2019 (Oral).
July	2018	Our paper PoseFlow is accepted in BMVC 2018.
Feb	2018	Our real-time accurate multi-person pose estimation system AlphaPose is released!

Project

AlphaPose
Real-Time and Accurate Multi-Person Pose Estimation & Tracking System

Publications

	COIN: Control-Inpainting Diffusion Prior for Human and Camera Motion Estimation Jiefeng Li, Ye Yuan, Davis Rempe, Haotian Zhang, Pavlo Molchanov, Jan Kautz, and Umar Iqbal ECCV, 2024 project page \| arXiv
	Denoising Vision Transformer Jiawei Yang, Katie Z Luo, Jiefeng Li, Congyue Deng, Leonidas J. Guibas, Dilip Krishnan, Kilian Q Weinberger, Yonglong Tian, and Yue Wang ECCV, 2024 (Oral Presentation) Project Page \| arXiv
	Balanced Parametric Body Prior for Implicit Clothed Human Reconstruction from a Monocular RGB Rong Xue, Jiefeng Li, and Cewu Lu IET Computer Vision, 2024 paper
	NIKI: Neural Inverse Kinematics with Invertible Neural Networks for 3D Human Pose and Shape Estimation Jiefeng Li, Siyuan Bian, Qi Liu, Jiasheng Tang, Fan Wang, and Cewu Lu CVPR, 2023 arXiv
	AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Time Hao-Shu Fang, Jiefeng Li, Hongyang Tang, Chao Xu, Haoyi Zhu, Yuliang Xiu, Yong-Lu Li, and Cewu Lu TPAMI, 2022 arXiv
	D&D: Learning Human Dynamics from Dynamic Camera Jiefeng Li, Siyuan Bian, Chao Xu, Gang Liu, Gang Yu, and Cewu Lu ECCV, 2022 (Oral Presentation - Top 2.7%)* arXiv
	Localization with Sampling-Argmax Jiefeng Li, Tong Chen, Ruiqi Shi, Yujing Lou, Yong-Lu Li, and Cewu Lu NeurIPS, 2021 project page \| arXiv \| zhihu
	Human Pose Regression with Residual Log-likelihood Estimation Jiefeng Li, Siyuan Bian, Ailing Zeng, Can Wang, Bo Pang, and Cewu Lu ICCV, 2021 (Oral Presentation - Top 3%) project page \| arXiv \| zhihu
	HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimation Jiefeng Li, Chao Xu, Zhicun Chen, Siyuan Bian, Lixin Yang, and Cewu Lu CVPR, 2021 project page \| arXiv \| zhihu
	HMOR: Hierarchical Multi-Person Ordinal Relations for Monocular Multi-Person 3D Pose Estimation Jiefeng Li, Can Wang, Wentao Liu, Chen Qian, and Cewu Lu ECCV, 2020 (Spotlight - Top 5%) arXiv
	CrowdPose: Efficient Crowded Scenes Pose Estimation and A New Benchmark Jiefeng Li, Can Wang, Hao Zhu, Yihuan Mao, Hao-Shu Fang, and Cewu Lu CVPR, 2019 (Oral - Top 5.6%) arXiv \| news
	Pose Flow: Efficient Online Pose Tracking Yuliang Xiu, Jiefeng Li, Haoyu Wang, Yinghong Fang, and Cewu Lu BMVC, 2018 arXiv
	Learning Analytical Posterior Probability for Human Mesh Recovery Qi Fang, Kang Chen, Yinghui Fan, Qing Shuai, Jiefeng Li, and Weidong Zhang CVPR, 2023 project page \| paper \| media coverage
	SmoothNet: A Plug-and-Play Network for Refining Human Poses in Videos Ailing Zeng, Lei Yang, Xuan Ju, Jiefeng Li, Jianyi Wang, and Qiang Xu ECCV, 2022 arXiv
	Constructing Balance from Imbalance for Long-tailed Image Recognition Yue Xu, Yong-Lu Li, Jiefeng Li, and Cewu Lu ECCV, 2022 arXiv
	ArtiBoost: Boosting Articulated 3D Hand-Object Pose Estimation via Online Exploration and Synthesis Lixin Yang, Kailin Li, Xinyu Zhan, Jun Lv, Wenqiang Xu, Jiefeng Li, and Cewu Lu CVPR, 2022 (Oral - Top 4%) arXiv
	Unified and Fast Human Trajectory Prediction Via Conditionally Parameterized Normalizing Flow Jianhua Sun, Zehao Wang, Jiefeng Li, and Cewu Lu RA-L, 2022 paper
	Correlation Field for Boosting 3D Object Detection in Structured Scenes Jianhua Sun, Hao-Shu Fang, Xianghui Zhu, Jiefeng Li, and Cewu Lu AAAI, 2022 paper
	Unsupervised Representation for Semantic Segmentation by Implicit Cycle-Attention Contrastive Learning Bo Pang, Yizhuo Li, Yifan Zhang, Gao Peng, Jiajun Tang, Jiefeng Li, and Cewu Lu AAAI, 2022 paper
	CPF: Learning a Contact Potential Field to Model the Hand-Object Interaction Lixin Yang, Xinyu Zhan, Kailin Li, Wenqiang Xu, Jiefeng Li, and Cewu Lu ICCV, 2021 project page \| arXiv \| zhihu
	TDAF: Top-Down Attention Framework for Vision Tasks Bo Pang, Yizhuo Li, Jiefeng Li, Muchen Li, Hanwen Cao, and Cewu Lu AAAI, 2020 arXiv
	Detailed 2D-3D Joint Representation for Human-Object Interaction Yong-Lu Li, Xinpeng Liu, Han Lu, Shiyi Wang, Junqi Liu, Jiefeng Li, and Cewu Lu CVPR, 2020 benchmark \| arXiv

Show all publications

Talks

Apr 3, 2024	Body Mesh Recovery with Inverse Kinematics – A new paradigm	ZongMu Technology
Dec 18, 2022	TechBeat: 多人场景，全身136关键点检测与跟踪框架AlphaPose技术讲解	TechBeat
Oct 19, 2022	From skeleton to body: Robust Human Pose Estimation	International Digital Economy Academy (IDEA)
Aug 24, 2022	Human Pose Regression with Residual Log-likelihood Estimation	Valse 2022
Jun 6, 2022	Regression From the Perspective of MLE	Artificial Intelligence Institute, SJTU
Sep 6, 2021	From skeleton to body: Keypoint Estimation is Helpful for Human Body Reconstruction	Max Planck Institute for Intelligent Systems (MPI-IS)
Sep 2, 2021	Human Pose Regression with Residual Log-likelihood Estimation	Tencent GY-Lab
July 28, 2021	HybrIK: 基于混合逆运动学解法的3D人体形态估计 \| CVPR 2021特别企划	智东西

Service

Conference Reviewer	NeurIPS, CVPR, ICCV, ECCV, IROS, AAAI
Journal Reviewer	IJCV, TIP, T-ASE, PR, JVIS

Template from Jon Barron's source code.