Haoxiang Li, Ph.D.

lhxustcer at gmail.com

Chief Scientist
Pixocial Technology
Bellevue WA, US

Blog
Github
Google Scholar
Mastodon

Researcher and engineer with expertise in Computer Vision, Deep Learning, and Robotics. My recent research focuses on image/video understanding and generation, 2D/3D perception and applying AI technologies to enhance and streamline business processes. Leading a team of scientists and engineers, I have gained experience across the entire product life cycle - from cultivation of ideas, prototyping, to research and development, and through to deployment and maintenance.

Recent

Internship opening: I am looking for motivated students to work at Pixocial AI Lab on Computer Vision, Deep Learning, and Generative AI.

  • 09/2024: Serve as Area Chair for WACV 2025
  • 08/2024: Serve as Senior PC for AAAI 2025
  • 07/2024: Our paper "UGG: Unified Generative Grasping" is accepted to ECCV 2024 as Oral. We can use diffusion model for dexterous grasping!
  • 05/2024: Our paper on 3D Shape Generation is accepted to ICML 2024
  • 07/2023: Our paper on IAE for 3D representation learning is accepted to ICCV 2023
  • 07/2023: Our paper on evidential modeling for visual recognition is accepted to ICCV 2023
  • 06/2023: Serve as Area Chair for WACV 2024
  • 11/2022: Our paper on Early-exiting dynamic neural networks (EDNN) is accepted to AAAI 2023, we formulate an EDNN as an additive model inspired by gradient boosting, and propose multiple training techniques to optimize the model effectively.
  • 11/2022: Serve as Panelist on Senior Member Review Panel
  • 10/2022: Serve as Area Chair for WACV 2023
  • 08/2022: Elevated to IEEE Senior Member
  • 07/2022: Our paper on Long-tailed Recognition is accepted to ECCV 2022, we introduced a sampling method to improve long-tailed recognition performance without extra computation: Breadcrumbs: Adversarial Class-Balanced Sampling for Long-tailed Recognition
  • 02/2022: Serve as Area Chair for ICPR 2022
  • 09/2021: Serve as Area Chair for CVPR 2022 and SPC for AAAI 2022
  • 07/2021: Serve as Area Chair for WACV 2022
  • 07/2021: Our paper on Long-tailed Recognition is accepted to ICCV 2021, we introduced GistNet to transfer geometric information from popular to low-shot classes.
  • 05/2021: Our paper on Human Pose Estimation and Tracking with GNN is accepted to CVPR 2021
  • 12/2020: Our paper on any-precision NN is accepted to AAAI 2021
  • 07/2020: One paper on the robotic wheelchair we built when I was doing my PhD is accepted to Transactions on Human-Robot Interaction (THRI)
  • 09/2019: Serve as area chair for WACV 2020
  • 08/2019: One paper on efforts towards interpretable face recognition is accepted to ICCV 2019 as oral!
  • 04/2018: The deep learning based face detection and recognition technology I made major contribution to at Adobe had been shipped in Adobe Photoshop, Lightroom, Premiere Pro! Feeling really happy.
  • 10/2017: Our face detection technology has been shipped in Photoshop October 2017 release!

Professional Services

Area Chair for

  • IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2022
  • IEEE Winter Conference on Applications of Computer Vision (WACV) 2020, 2022, 2023, 2024, 2025

Senior PC for

  • The AAAI Conference on Artificial Intelligence (AAAI) 2022, 2025

IEEE Senior Member

Publications

Up-to-date full list on Google Scholar

Computer Vision - 3D Vision

Enhancing Implicit Shape Generators Using Topological Regularizations,

Chen, Liyan and Zheng, Yan and Li, Yang and Jagarapu, Lohit Anirudh and Li, Haoxiang and Kang, Hao and Hua, Gang and Huang, Qixing, International Conference on Machine Learning (ICML), 2024 [Paper]

Implicit Autoencoder for Point-Cloud Self-Supervised Representation Learning,

Yan, Siming and Yang, Zhenpei and Li, Haoxiang and Song, Chen and Guan, Li and Kang, Hao and Hua, Gang and Huang, Qixing, IEEE/CVF International Conference on Computer Vision (ICCV), 2023 [Paper]

Computer Vision - Recognition and Representation

Flexible Visual Recognition by Evidential Modeling of Confusion and Ignorance,

Fan, Lei and Liu, Bo and Li, Haoxiang and Wu, Ying and Hua, Gang, IEEE/CVF International Conference on Computer Vision (ICCV), 2023 [Paper]

Self-supervised pretraining with classification labels for temporal activity detection,

Kahatapitiya, Kumara and Ren, Zhou and Li, Haoxiang and Wu, Zhenyu and Ryoo, Michael S, AAAI Conference on Artificial Intelligence (AAAI), 2023 [Paper]

Breadcrumbs: Adversarial class-balanced sampling for long-tailed recognition,

Liu, Bo and Li, Haoxiang and Kang, Hao and Hua, Gang and Vasconcelos, Nuno, European Conference on Computer Vision (ECCV), 2022 [Paper]

Gistnet: a geometric structure transfer network for long-tailed recognition,

Liu, Bo and Li, Haoxiang and Kang, Hao and Hua, Gang and Vasconcelos, Nuno, Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2021 [Paper]

Few-shot open-set recognition using meta-learning,

Liu, Bo and Kang, Hao and Li, Haoxiang and Hua, Gang and Vasconcelos, Nuno, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020 [Paper]

A modulation module for multi-task learning with applications in image retrieval,

Zhao, Xiangyun and Li, Haoxiang and Shen, Xiaohui and Liang, Xiaodan and Wu, Ying, Proceedings of the European Conference on Computer Vision (ECCV), 2018 [Paper]

Contemplating visual emotions: Understanding and overcoming dataset bias,

Panda, Rameswar and Zhang, Jianming and Li, Haoxiang and Lee, Joon-Young and Lu, Xin and Roy-Chowdhury, Amit K, Proceedings of the European Conference on Computer Vision (ECCV), 2018 [Paper]

Vqs: Linking segmentations to questions and answers for supervised attention in vqa and question-focused semantic segmentation,

Gan, Chuang and Li, Yandong and Li, Haoxiang and Sun, Chen and Gong, Boqing, Proceedings of the IEEE international conference on computer vision (ICCV), 2017 [Paper]

Computer Vision - Human and Face Understanding

Learning dynamics via graph neural networks for human pose estimation and tracking,

Yang, Yiding and Ren, Zhou and Li, Haoxiang and Zhou, Chunluan and Wang, Xinchao and Hua, Gang, Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021 [Paper]

Towards interpretable face recognition,

Yin, Bangjie and Tran, Luan and Li, Haoxiang and Shen, Xiaohui and Liu, Xiaoming, Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2019 [Paper]

Deep face detector adaptation without negative transfer or catastrophic forgetting,

Jamal, Muhammad Abdullah and Li, Haoxiang and Gong, Boqing, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018 [Paper]

Learning dense facial correspondences in unconstrained images,

Yu, Ronald and Saito, Shunsuke and Li, Haoxiang and Ceylan, Duygu and Li, Hao, Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2017 [Paper]

Probabilistic elastic part model: a pose-invariant representation for real-world face verification,

Li, Haoxiang and Hua, Gang, IEEE transactions on pattern analysis and machine intelligence (T-PAMI), 2017 [Paper]

A multi-level contextual model for person recognition in photo albums,

Li, Haoxiang and Brandt, Jonathan and Lin, Zhe and Shen, Xiaohui and Hua, Gang, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016 [Paper]

Labeled faces in the wild: A survey,

Learned-Miller, Erik and Huang, Gary B and RoyChowdhury, Aruni and Li, Haoxiang and Hua, Gang, Advances in face detection and facial image analysis, Book Chapter (Book), 2016 [Paper]

Report on the FG 2015 video person recognition evaluation,

Beveridge, J Ross and Zhang, Hao and Draper, Bruce A and Flynn, Patrick J and Feng, Zhenhua and Huber, Patrik and Kittler, Josef and Huang, Zhiwu and Li, Shaoxin and Li, Yan and others, 11th IEEE international conference and workshops on Automatic Face and Gesture Recognition (FG), 2015 [Paper]

Hierarchical-pep model for real-world face recognition,

Li, Haoxiang and Hua, Gang, Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), 2015 [Paper]

A convolutional neural network cascade for face detection,

Li, Haoxiang and Lin, Zhe and Shen, Xiaohui and Brandt, Jonathan and Hua, Gang, Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), 2015 [Paper]

Efficient boosted exemplar-based face detection,

Li, Haoxiang and Lin, Zhe and Brandt, Jonathan and Shen, Xiaohui and Hua, Gang, Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), 2014 [Paper]

The IJCB 2014 PaSC Video Face and Person Recognition Competition,

Beveridge, J. Ross and Zhang, Hao and Flynn, Patrick J. and Lee, Yooyoung and Liong, Venice Erin and Lu, Jiwen and de Assis Angeloni, Marcus and de Freitas Pereira, Tiago and Li, Haoxiang and Hua, Gang and others, IEEE International Joint Conference on Biometrics (IJCB), 2014 [Paper]

Eigen-Pep for Video Face Recognition,

Li, Haoxiang and Hua, Gang and Shen, Xiaohui and Lin, Zhe and Brandt, Jonathan, Asian Conference on Computer Vision (ACCV), 2014 [Paper]

Probabilistic elastic matching for pose variant face verification,

Li, Haoxiang and Hua, Gang and Lin, Zhe and Brandt, Jonathan and Yang, Jianchao, Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR), 2013 [Paper]

Robotics

UGG: Unified Generative Grasping, (Oral Presentation)

Lu, Jiaxin and Kang, Hao and Li, Haoxiang and Liu, Bo and Yang, Yiding and Huang, Qixing and Hua, Gang, Proceedings of the European Conference on Computer Vision (ECCV), 2024 [Paper]

Egocentric Computer Vision for Hands-Free Robotic Wheelchair Navigation,

Kutbi, Mohammed and Li, Haoxiang and Chang, Yizhe and Sun, Bo and Li, Xin and Cai, Changjiang and Agadakos, Nikolaos and Hua, Gang and Mordohai, Philippos, Journal of Intelligent & Robotic Systems (JIRS), 2023 [Paper]

Usability studies of an egocentric vision-based robotic wheelchair,

Kutbi, Mohammed and Du, Xiaoxue and Chang, Yizhe and Sun, Bo and Agadakos, Nikolaos and Li, Haoxiang and Hua, Gang and Mordohai, Philippos, ACM Transactions on Human-Robot Interaction (THRI), 2020 [Paper]

Flycam: Multitouch gesture controlled drone gimbal photography,

Kang, Hao and Li, Haoxiang and Zhang, Jianming and Lu, Xin and Benes, Bedrich, IEEE Robotics and Automation Letters (RAL), 2018 [Paper]

Active object perceiver: Recognition-guided policy learning for object searching on mobile robots,

Ye, Xin and Lin, Zhe and Li, Haoxiang and Zheng, Shibin and Yang, Yezhou, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2018 [Paper]

An egocentric computer vision based co-robot wheelchair,

Li, Haoxiang and Kutbi, Mohammed and Li, Xin and Cai, Changjiang and Mordohai, Philippos and Hua, Gang, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2016 [Paper]

Egocentric Computer Vision based Wheelchair Robot Control,

Li, Haoxiang and Hua, Gang, ICRA Late Breaking Results Poster Session (ICRA), 2015 [Paper]

Efficient Deep Learning

Boosted Dynamic Neural Networks,

Yu, Haichao and Li, Haoxiang and Hua, Gang and Huang, Gao and Shi, Humphrey, AAAI Conference on Artificial Intelligence (AAAI), 2023 [Paper]

Any-precision deep neural networks,

Yu, Haichao and Li, Haoxiang and Shi, Humphrey and Huang, Thomas S and Hua, Gang, Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2021 [Paper]