Ailing Zeng 曾爱玲

Ph.D. Department of Computer Science and Engineering The Chinese University of Hong Kong Visiting Scholar Robotics Institute, Carnegie Mellon University Google Scholar Resume

Biography

I obtained my Ph.D. degree from the department of computer science and engineering, the Chinese University of Hong Kong, supervised by Prof. Qiang Xu. I'm working closely with Prof. Kris Kitani and Prof. László A. Jeni as a visiting scholar in the Robotics Institute, Carnegie Mellon University. I have also spent wonderful time at Baidu, Microsoft Asia, and Sensetime as a research intern. Before that I got my bachelor’s degrees (Top 1% graduates) of Electronic Engineering and Finance mathematics in Xiamen University.

My research interests include: human-centric visual understanding by deep learning, like 3d human pose estimation/reconstruction/recognition/prediction/synthesis, and general time series analysis.

Looking for self-motivated interns. Please drop your CV to me if you are interested in.

News

• [2022.07] Three papers have been accepted to ECCV 2022! Cheers!

• [2022.01] T-WaveNet has been accepted to ICLR 2022!

• [2021.07] Two papers accepted to ICCV 2021! (1 oral and 1 poster)

• [2021.05] One paper accepted to IJCAI 2021!

• [2020.07] One paper accepted to ECCV 2020!

Publications

DeciWatch: A Simple Baseline for 10× Efficient 2D and 3D Pose Estimation [Website] [Code] [Data] Ailing Zeng, Xuan Ju, Lei Yang, Ruiyuan Gao, Xizhou Zhu, Bo Dai, Qiang Xu European Conference on Computer Vision (ECCV), 2022 DeciWatch is a simple baseline framework with the sample-denoise-recover scheme for video-based 2D/3D human pose estimation that can achieve 10 times efficiency improvement over existing works without any performance degradation.

SmoothNet: A Plug-and-Play Network for Refining Human Poses in Videos [Website] [Code] Ailing Zeng, Lei Yang, Xuan Ju, Jiefeng Li, Jianyi Wang, Qiang Xu European Conference on Computer Vision (ECCV), 2022 SmoothNet is a refinement network, which can be attached to any existing pose estimators to improve their temporal smoothness and enhance the per-frame precision.

HuMMan: Multi-Modal 4D Human Dataset for Versatile Sensing and Modeling [Website] Zhongang Cai*, Daxuan Ren*, Ailing Zeng*, Zhengyu Lin*, Tao Yu*, Wenjia Wang*, Xiangyu Fan, Yang Gao, Yifan Yu, Liang Pan, Fangzhou Hong, Mingyuan Zhang, Chen Change Loy, Lei Yang^, Ziwei Liu^ European Conference on Computer Vision (ECCV), 2022 HuMMan is a large-scale multi-modal 4D human dataset with 1000 human subjects, 400k sequences and 60M frames.

Are Transformers Effective for Time Series Forecasting? [Code] [知乎] Ailing Zeng, Muxi Chen, Lei Zhang, Qiang Xu arXiv preprint, 2022 We try to question the validity of Transformer-based time series forecasting solutions via a linear model.

T-WaveNet: Tree-Structured Wavelet Neural Network for Sensor-Based Time Series Analysis [Code] Minhao Liu, Ailing Zeng, Qiuxia Lai, Qiang Xu 2022 Tenth International Conference on Learning Representations (ICLR'22) We introduce a tree-structured wavelet deep neural network to effectively extract more discriminative and expressive feature representations in time series signals.

Learning Skeletal Graph Neural Networks for Hard 3D Pose Estimation [Code] Ailing Zeng, Xiao Sun, Lei Yang, Nanxuan Zhao, Minhao Liu, Qiang Xu 2021 IEEE/CVF International Conference on Computer Vision (ICCV’21) We explore a novel skeletal graph neural network to deal with hard poses in 3D pose estimation.

Human Pose Regression with Residual Log-likelihood Estimation [Code] [知乎] Jiefeng Li, Siyuan Bian, Ailing Zeng, Can Wang, Bo Pang, Wentao Liu and Cewu Lu (Oral) 2021 IEEE/CVF International Conference on Computer Vision (ICCV’21) We propose a novel regression paradigm, which is superior to the heatmap-based methods for the first time, especially on multi-person pose estimation.

Relational Graph Neural Network Design via Progressive Neural Architecture Search [Code] Ailing Zeng, Minhao Liu, Zhiwei Liu, Ruiyuan Gao, Qiang Xu arXiv preprint, 2021 We propose a ladder-style GNN aggregation scheme to handle with low homophily nodes on semi-supervised node classification.

SRNet: Improving Generalization in 3D Human Pose Estimation with a Split-and-Recombine Approach [Code] Ailing Zeng, Xiao Sun, Fuyang Huang, Minhao Liu, Qiang Xu, Stephen Lin 2020 European Conference on Computer Vision (ECCV’20) We design a split-and-recombine approach to improve generalization performance in 3D human pose estimation, especially on rare and unseen poses.

Skimming and Scanning for Efficient Action Recognition in Untrimmed Videos Yunyan Hong*, Ailing Zeng*, Min Li, Cewu Lu, Li Jiang, Qiang Xu 2021 The International Congress on Image and Signal Processing, BioMedical Engineering and Informatics We propose a highly efficient and effective clip-level framework based on skim-scan techniques for untrimmed video action recognition.

Time Series is a Special Sequence: Forecasting with Sample Convolution and Interaction [Code] Minhao Liu, Ailing Zeng, Qiuxia Lai, Qiang Xu arXiv preprint, 2021 We propose a novel and general CNN-based backbone for the time series forecasting problem, achieving SOTAs on several real-world datasets.

Information Bottleneck Approach to Spatial Attention Learning Qiuxia Lai, Yu Li, Ailing Zeng, Minhao Liu, Qiang Xu, Hanqiu Sun 2021 The 30th International Joint Conference on Artificial Intelligence (IJCAI’21) We propose a spatial attention mechanism based on IB theory for generating probabilistic maps to help visual recognition tasks.

DeepFuse: An IMU-Aware Network for Real-Time 3D Human Pose Estimation from Multi-View Image Fuyang Huang, Ailing Zeng, Minhao Liu, Qiang Xu 2020 IEEE Winter Conference on Applications of Computer Vision (WACV’20) We propose a two-stage 3D network to estimate 3D human pose by fusing body-worn Inertial Measurement Unit (IMU) data and multi-view images deeply.

Differentiable Kernel Evolution Yu Liu, Jihao Liu, Ailing Zeng, Xiaogang Wang 2019 IEEE/CVF International Conference on Computer Vision (ICCV’19) We propose a differentiable kernel evolution (DKE) algorithm to find a better layer-operator for CNNs.

Structure-Aware 3D Hourglass Network for Hand Pose Estimation from Single Depth Image Fuyang Huang, Ailing Zeng, Minhao Liu, Jing Qin, Qiang Xu 2018 The British Machine Vision Conference (BMVC’18) We propose a structure-aware 3D hourglass network for hand pose estimation from a single depth image.

Working Experience

• Research intern at Sensetime Research, X-lab. (Nov. 2020 to Apr. 2022)
  Worked on robust and efficient 2D/3D human pose estimation&reconstruction from videos.

• Research intern at Microsoft Asia Research, Visual Computing Group. (Dec. 2019 to Sep. 2020)
  Worked on improving the generalization ability of single-view 3D human pose estimation from either images or videos.

• Research intern at Baidu Research. (June 2018 to Sep. 2018)
  Worked on 2D human pose estimation.

• TMT investment intern at ASB Ventures. (March 2017 to Aug. 2017)
  Analyzed technological feasibility of start-up companies.

• In recent years, I am a reviewer of the following conferences and journals: CVPR, ICCV, ICML, NeurIPS, AAAI, IJCV.

Honors & Awards

• Full Postgraduate Studentship, CUHK (2017 - 2021)

• Excellent League Member, Top 1% student of Xiamen University

• National Scholarship (2015)