Stay active, keep calm, and carry on.
JUNFU PU  

Researcher at Tencent, Ph.D

Dept. Electronic Engineering and Information Science (EEIS)

University of Science and Technology of China (USTC)

Email: pjh (at) mail.ustc.edu.cn

View on    


News Open-Resources Publications Projects Experience Teaching Link

Biography

I am a researcher at Tencent ARC Lab, specializing in computer vision and multimedia applications. My work focuses on multimedia understanding and search, as well as developing multimodal large language models (LLMs) for vision-language applications. I got my PhD degree from University of Science and Technology of China (USTC), adviced by Prof. Wengang Zhou and Prof. Houqiang Li. Before that, I received the B.E. degree in EE from School of the Gifted Young, University of Science and Technology of China (USTC) in 2015. I was a research intern at Satoh Lab, National Institute of Informatics (NII) from Aug. 2016 to Nov. 2016. My research project at NII is video event retrieval with the supervision of Prof. Shin'ichi Satoh.

During my PhD period, my research interests include image processing and computer vision. Specifically, I focus on sign language recognition and translation. Besides, I pay attention to sign video generation task as well.

If you are interested in our research on sign language recognition, please visit the homepage of our Visual Sign Language Research Group to access our Chinese SLR dataset and research papers.

News   [Top]

  • 2024: Three papers accepted to CVPR, ACM MM, and TCSVT.
  • 2022 & 2023: Three papers accepted to IEEE Transactions on Multimedia (TMM).
  • Jul. 2020: One paper is accepted to ACM MM 2020.
  • Oct. 2019: Our paper is selected as Best Student Paper Finalists for ICIP 2019.
  • Feb. 2019: One paper is accepted to CVPR 2019.
  • Apr. 2018: One paper is accepted to IJCAI-ECAI 2018.
  • Apr. 2017: One paper is accepted to ICIP 2017.
  • Oct. 2016: Awarded National Scholarship.
  • Jun. 2016: Application for Internship Program is accepted by NII.
  • Jun. 2016: One paper is accepted to PCM 2016.
  • Mar. 2016: One paper (co-author) is accepted to ICME 2016.
  • Sep. 2015: One paper is accepted to MMM 2016.
  • Open Resources  [Top]

    We have released some resources for research purpose about Sign Language Recognition (SLR). Please refer to the following links:

    Source code for project or paper:

    Publications  [Top]

    Selected Publications [see full publication list...]
    SynopGround: A Large-Scale Dataset for Multi-Paragraph Video Grounding from TV Dramas and Synopses
    Chaolei Tan, Zihang Lin, Junfu Pu, Zhongang Qi, Wei-Yi Pei, Zhi Qu, Yexin Wang, Ying Shan, Wei-Shi Zheng, and Jian-Fang Hu
    ACM International Conference on Multimedia (ACM MM), 2024
    [Paper] [BibTex] [Project Page]
    Learning Music-Dance Representations through Explicit-Implicit Rhythm Synchronization
    Jiashuo Yu, Junfu Pu✉️, Ying Cheng, Rui Feng, and Ying Shan (✉️corresponding author)
    IEEE Transactions on Multimedia (TMM), 2023
    [Paper (arXiv)] [BibTex]
    Prior-aware Cross Modality Augmentation Learning for Continuous Sign Language Recognition
    Hezhen Hu, Junfu Pu, Wengang Zhou, and Houqiang Li
    IEEE Transactions on Multimedia (TMM), 2023
    [Paper] [BibTex]
    Collaborative Multilingual Continuous Sign Language Recognition: A Unified Framework
    Hezhen Hu*, Junfu Pu*, Wengang Zhou, and Houqiang Li (*equal contribution)
    IEEE Transactions on Multimedia (TMM), 2022
    [Paper] [BibTex]
    PC-Dance: Posture-controllable Music-driven Dance Synthesis
    Jibin Gao, Junfu Pu, Honglun Zhang, Ying Shan, and Wei-Shi Zheng
    ACM International Conference on Multimedia (ACM MM), 2022
    [Paper] [BibTex]
    Improving Sign Language Translation with Monolingual Data by Sign Back-Translation
    Hao Zhou, Wengang Zhou, Weizhen Qi, Junfu Pu, and Houqiang Li
    IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021
    [Paper (arXiv)] [BibTex]
    Boosting Continuous Sign Language Recognition via Cross Modality Augmentation
    Junfu Pu, Wengang Zhou, Hezhen Hu, and Houqiang Li
    ACM International Conference on Multimedia (ACM MM), 2020
    [Paper] [BibTex] [arXiv] [Presentation]
    Global-local Enhancement Network for NMFs-aware Sign Language Recognition
    Hezhen Hu, Wengang Zhou, Junfu Pu, and Houqiang Li
    ACM Transactions on Multimedia Computing Communications and Applications (TOMM), 2020
    [Paper] [BibTex] [arXiv]
    Iterative Alignment Network for Continuous Sign Language Recognition
    Junfu Pu, Wengang Zhou, and Houqiang Li
    IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019
    [Paper] [BibTex] [Poster] [Project]
    Continuous Sign Language Recognition via Reinforcement Learning
    Zhihao Zhang, Junfu Pu, Liansheng Zhuang, Wengang Zhou, and Houqiang Li
    International Conference on Image Processing (ICIP, Best Student Paper Finalists), 2019
    [Paper] [BibTex] [Best Student Paper Finalists]
    Dilated Convolutional Network with Iterative Optimization for Coutinuous Sign Language Recognition
    Junfu Pu, Wengang Zhou, and Houqiang Li
    International Joint Conference on Artificial Intelligence (IJCAI), 2018
    [Paper] [BibTex] [Slides] [Code]
    Deep Supervised Quantization by Self-Organizing Map
    Min Wang, Wengang Zhou, Qi Tian, Junfu Pu, and Houqiang Li
    ACM International Conference on Multimedia (ACM MM), 2017
    [Paper] [BibTex]
    Chinese Sign Language Recognition with Adaptive HMM
    Jihai Zhang, Wengang Zhou, Chao Xie, Junfu Pu, and Houqiang Li
    International Conference on Multimedia and Expo (ICME), 2016
    [Paper] [BibTex]

    Patents
    1. 于家硕, 蒲俊福, 单瀛. 一种模型训练方法、装置、设备和存储介质. 202210148011.9. 2022.02.17申请.
    2. 李厚强, 周文罡, 胡鹤臻, 蒲俊福. 基于全局-局部特征增强的孤立词手语识别方法及系统. 202010513333.X. 2020.06.08申请. 2022.07.15授权.
    3. 李厚强, 周文罡, 胡鹤臻, 蒲俊福. 一种基于跨模态数据增广的连续手语识别方法. 202011060274.1. 2020.09.30申请. 2022.07.15授权.
    4. 李厚强, 周文罡, 蒲俊福, 胡鹤臻. 基于多语言协同的连续手语识别系统. 202011060272.2. 2020.09.30申请. 2022.07.15授权.
    5. 李厚强, 周文罡, 蒲俊福. 手语识别方法及装置. 201910456373.2. 2019.05.29申请. 2021.07.06授权.
    6. 李厚强, 蒲俊福, 周文罡. 一种手语识别方法及装置. 201810743921.5. 2018.07.09申请.

    Projects  [Top]

    DeepCraft: Fall in Love with the World
    Yunfeng Wang, Ke Sun, Yue Lv, and Junfu Pu
    Project for hackathon (HACKxSJTU), 2017.
    Finished at SJTU, Shanghai, China
    [Project] [Code] [Prize] [Certification]
    CaptchaLess: A Chrome Extension for Auto-filling Captchas
    Yunfeng Wang, Zhihua Huang, and Junfu Pu
    Final Project for Video Technology Course, Autumn, 2015
    Finished at University of Science and Technology of China (USTC), China
    [Project] [Code] [Chrome Extension] [Slides]

    Experience  [Top]

    Teaching Assistant  [Top]



    © Junfu Pu 2024       Last updated on Aug 2024