Hongtao Wen (文洪涛)

HWen_smile.png

Ph.D. student at the vLAR group, PolyU.

I am a Ph.D. student in the Visual Learning and Reasoning (vLAR) Group at The Hong Kong Polytechnic University (PolyU), under the supervision of Prof. Bo Yang. Before that, I received my M.Eng. and B.Eng. degrees from Dalian University of Technology (DUT), where I was supervised by Prof. Yi Sun. I graduated with my B.Eng. degree in 2020 with the honor of Outstanding Graduate of Liaoning Province.

I work in robotics, computer vision, and deep learning. My current research focuses on dynamic manipulation and robot learning.

  News

Aug, 2023 I joined the vLAR group, PolyU as a Ph.D. student.
Apr, 2023 We got the 2nd place in ICDAR 2023 Competition on Hierarchical Text Detection and Recognition.
We beat Amazon, Alibaba, Nvidia, Huawei, etc.
Jan, 2023 I joined the DeepSE Lab, HKUST as a Research Assistant,
supervised by Prof. Sung Kim and Dr. Sungrae Park.

  Selected Publications

2026

  1. PhysInOne.png
    PhysInOne: Visual Physics Learning and Reasoning in One Suite
    Siyuan ZhouHejun WangHu ChengJinxi Li, Dongsheng Wang, Junwei Jiang, Yixiao Jin, Jiayue Huang, Shiwei Mao, Shangjia Liu, Yafei Yang, Hongkang Song, Shenxing WeiZihui Zhang, Peng Huang, Shijie Liu, Zhengli Hao, Hao Li, Yitian Li, Wenqi Zhou, Zhihan Zhao, Zongqi He, Hongtao Wen, Shouwang Huang, Peng YunBowen Cheng, Pok Kazaf Fu, Wai Kit Lai, Jiahao Chen, Kaiyuan Wang, Zhixuan Sun, Ziqi Li, Haochen Hu, Di Zhang, Chun Ho YuenBing Wang, Zhihua Wang, Chuhang Zou, and Bo Yang
    In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2026

2025

  1. LogoSP.gif
    LogoSP: Local-global Grouping of Superpoints for Unsupervised Semantic Segmentation of 3D Point Clouds
    Zihui Zhang, Weisheng Dai, Hongtao Wen, and Bo Yang
    In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2025
  2. GrabS.gif
    GrabS: Generative Embodied Agent for 3D Object Segmentation without Scene Supervision
    Zihui ZhangYafei YangHongtao Wen, and Bo Yang
    In The Thirteenth International Conference on Learning Representations, 2025
    Spotlight

2022

  1. TransGrasp.png
    TransGrasp: Grasp pose estimation of a category of objects by transferring grasps from only one labeled instance
    Hongtao WenJianhang YanWanli Peng, and Yi Sun
    In European Conference on Computer Vision, 2022
  2. SSC-6D.jpg
    Self-supervised category-level 6D object pose estimation with deep implicit shape representation
    Wanli PengJianhang YanHongtao Wen, and Yi Sun
    In Proceedings of the AAAI Conference on Artificial Intelligence, 2022
In my daily life, I excel in swimming (fourth place in the DUT breaststroke final) and love climbing mountains for a broader perspective. It may come as a surprise that I participated in Chinese folk dance for the DUT art performance (Fenglan Cup, Video). Feel free to reach out to me via email (ht.wen@outlook.com).