as Miao Xin's Homepage
Miao
≡
profile photo

Miao Xin

I am an associate research fellow at Institute of Automation, Chinese Academy of Sciences (CASIA), Beijing, where I work on embodied human-robot interaction and multi-modal decision intelligence. Before that, I did my PhD at Beihang University and Harvard University.

Email  /  Google Scholar  / 

News

  • 01/2024, our paper has been accepted by ICASSP.

  • 02/2023, our paper has been accepted by IEEE TMM.

  • 01/2022, I won the Golden Prize in the 1st China Postdoctoral Innovation & Entrepreneurship Competition.

  • 10/2021, one paper has been accepted by ACM MM 2021.

  • 07/2021, our CVPR-W papers achieved two Workshop Best Paper Awards.

    • Recent Projects

  • Virtual Life: Autonomous Agents in Metaverse

    • This project aims to create 3D digital humans with autonomous behavior in the virtual world. They can live autonomously in virtual worlds, acting as the brains of service robots, and driving them to serve others (elderly and Alzheimer's patients) in real environments. We built a real indoor environment equipped with cameras, microphones, and sensors. The system continuously records all behaviors and interactions of the subjects. This information will be mapped to the virtual world environment. We employ reinforcement learning techniques to train the agents of virtual humans to predict the intentions of physical people. Our goal is to make them capable of decision-making and reasoning.

    Sponsored by: National Natural Science Foundation of China (NSFC).

    Research

    1. We Choose to Go to Space: Agent-driven Human and Multi-Robot Collaboration in Microgravity
      Miao Xin*, Zhongrui You, Zihan Zhang, Taoran Jiang, Tingjia Xu, Haotian Liang, Guojing Ge, Yuchen Ji, Shentong Mo, Jian Cheng
      paper | video | project page

    2. An Intelligent Fault Diagnosis Framework for Rolling Bearings with Integrated Feature Extraction and Ordering-based Causal Discovery
      Xu Ding, Junlong Wang, Hao Wu, Juan Xu, Miao Xin*
      IEEE Sensors Journal 2024 | paper

    3. Causal Discovery for Rolling Bearing Fault under Missing Data: From the Perspective of Causal Effect and Information Flow
      Xu Ding, Hao Wu, Junlong Wang, Miao Xin*
      IEEE Transactions on Instrumentation & Measurement (TIM) 2024

    4. Tree of Uncertain Thoughts Reasoning for Large Language Models
      Shentong Mo, Miao Xin*
      ICASSP 2024 | paper

    5. BFRFormer: Transformer-based Generator for Real-world Blind Face Restoration
      Guojing Ge, Qi Song, Guibo Zhu, Yuting Zhang, Jinglu Chen, Miao Xin, Ming Tang, Jinqiao Wang
      ICASSP 2024 | paper

    6. BSTG-Trans: A Bayesian Spatial-Temporal Graph Transformer for Long-term Pose Forecasting
      Shentong Mo, Miao Xin*
      IEEE Transactions on Multimedia (TMM) 2023 | paper

    7. OsGG-Net: One-step Graph Generation Network for Unbiased Head Pose Estimation
      Shentong Mo, Miao Xin*
      ACM MM 2021 | paper | code

    8. EVA-GCN: Head Pose Estimation Based on Graph Convolutional Networks
      Miao Xin*, Shentong Mo, Yuanze Lin
      CVPR 2021 Workshop | paper | project page
      AMFG Workshop Best Paper

    9. Long-Term Head Pose Forecasting Conditioned on the Gaze-Guiding Prior
      Shentong Mo, Miao Xin*
      CVPR 2021 Workshop | paper
      Precognition Workshop Best Paper

    10. Entanglement Loss for Context-Based Still Image Action Recognition
      Miao Xin, Shuhang Wang, Jian Cheng
      ICME 2019 | paper

    11. End-to-end Temporal Attention Extraction and Human Action Recognition
      Hong Zhang1, Miao Xin1, Shuhang Wang, Yifan Yang, Lei Zhang
      Machine Vision and Applications 2018 | paper

    12. Learning Discriminative Action and Context Representations for Action Recognition in Still Images
      Miao Xin, Hong Zhang, Ding Yuan, Mingui Sun
      ICME 2017 | paper

    13. Recurrent Temporal Sparse Autoencoder for attention-based action recognition
      Miao Xin, Hong Zhang, Mingui Sun, Ding Yuan
      IJCNN 2016 | paper

    14. ARCH: Adaptive recurrent-convolutional hybrid networks for long-term action recognition
      Miao Xin, Hong Zhang, Helong Wang, Mingui Sun, Ding Yuan
      Neurocomputing 2016 | paper

    Services

  • Expert group member of Chinese Association of Automation（CAA）.

  • Co-advisor for postgraduate students of Beihang University and Hefei University of Technology.

  • Journal Reviewers of Neurocomputing.

  • Conference Reviewers of CVPR 2022, CVPR 2021, AAAI 2021, CVPR 2020, AAAI 2020, ICME 2020, ICME 2018, etc.

    • Awards

  • 2022, Golden Prize in the 1st China Postdoctoral Innovation & Entrepreneurship Competition.

  • 2017, National Scholarship for Graduate Students