Xuanbai Chen

Xuanbai Chen is an Applied Scientist in AWS AI Rekognition team, and is currently working on face recognition and trustworthy AI.

Prior to AWS, he obtained my Master of Science in Computer Vision (MSCV) Degree in the Robotics Institute, School of Computer Science, Carnegie Mellon University. He worked with Prof. Fernando De la Torre on Fairness for Generation Model and 3D Face modeling. He obtained my Bachelor of Engineering in Computer Science (B.Eng in CS) Degree in College of Computer Science, Nankai University, working on Video Summarization.

Email  /  LinkedIn  /  Google Scholar  /  CV

profile photo
News

  • 2024-03: Our paper for video summarization has been accepted by KBS.
  • 2024-02: Our paper in better generating human images has been accepted by CVPR 2024.
  • 2023-10: ITI-GEN has entered the Best Paper Finalist (17 out of 2161 accepted paper Top 0.7%, 8260 submissions Top 0.2%) for ICCV 2023.
  • 2023-08: Our ITI-GEN has been accepted by ICCV 2023 as an Oral paper.
  • 2023-07: I joined AWS AI Rekognition team as an Applied Scientist!
  • Publications (* indicates equal contribution)
            Video Summarization via Knowledge-Aware Multimodal Deep Networks
    Jiehang Xie, Xuanbai Chen, Sicheng Zhao, Shao-Ping Lu
    KBS, 2024
    project page

    Propose a novel video summarization approach based on a knowledge-aware multimodal network.

            Towards Effective Usage of Human-Centric Priors in Diffusion Models for Text-based Human Image Generation
    Junyan Wang, Zhenhong Sun, Zhiyu Tan, Xuanbai Chen, Weihua Chen, Hao Li, Cheng Zhang, Yang Song
    CVPR, 2024
    project page / arXiv / code

    Integrating human-centric priors to help t2i model better generate human images.

            ITI-GEN: Inclusive Text-to-Image Generation
    Cheng Zhang, Xuanbai Chen, Siqi Chai, Chen Henry Wu, Dmitry Lagun, Thabo Beeler, Fernando De la Torre
    ICCV, 2023   (Oral Presentation, Best Paper Finalist, 17 out of 8260 submissions)
    project page / arXiv / code / data / poster

    Debias the text-to-image generation model and implement the inclusiveness.

            Controllable 3D Generative Adversarial Face Model via Disentangling Shape and Appearance
    Fariborz Taherkhani, Aashish Rai, Quankai Gao, Shaunak Srivastava, Xuanbai Chen, Fernando De la Torre, Steven Song, Aayush Prakash, Daeil Kim
    WACV, 2023
    project page / arXiv / code / video

    A new 3D face generative model that can decouple identity and expression and provides granular control over expressions.

            A Knowledge Augmented and Multimodal-Based Framework for Video Summarization
    Jiehang Xie*, Xuanbai Chen*, Shao-Ping Lu, Yulu Yang
    ACMMM, 2022
    project page / video

    Propose a knowledge augmented and multimodal-based video summarization method, by considering multichannel information and using the impact of external knowledge.

            Multimodal-based and Aesthetic-guided Narrative Video Summarization
    Jiehang Xie, Xuanbai Chen, Tianyi Zhang, Yixuan Zhang, Shao-Ping Lu, Pablo Cesar, Yulu Yang
    IEEE TMM, 2022
    project page

    Introduce a multimodal-based and aesthetic-guided narrative video summarization method, by leveraging multimodal information through specified key shots selection, subtitle summarization, and highlight extraction components.

            Emotional Semantics-Preserved and Feature-Aligned CycleGAN for Visual Emotion Adaptation
    Sicheng Zhao*, Xuanbai Chen*, Xiangyu Yue, Chuang Lin, Pengfei Xu, Ravi Krishna, Jufeng Yang, Ding Guiguang, Alberto Sangiovanni Vincentelli, Kurt Keutzer
    IEEE TCYB, 2021
    project page / arxiv

    Design a novel end-to-end cycle-consistent adversarial model, to focus on UDA in visual emotion analysis for both emotion distribution learning and dominant emotion classification.

    Education and Experiences

    AWS AI Rekognition, Jul. 2023 - Present

    Applied Scientist

    Carnegie Mellon University, Jan. 2022 - May. 2023

    Master of Science in Computer Vision

    AWS AI Rekognition, May. 2022 - Aug. 2022

    Applied Scientist Intern

    Sensetime, Aug. 2021 - Dec. 2021

    Research Engineer Intern

    University of California, Berkeley, Jan. 2020 - May. 2020

    Exchange Program

    Nankai University, Sep. 2017 - Jun. 2021

    Bachelor of Engineering in Computer Science

    Academic Services

  • Journal Reviewer: IEEE TCYB, TMM, TNSRE, KBS
  • Conference Reviewer: ACMMM 2023 2024, CVPR 2024, ECCV 2024
  • Student Volunteer: ICCV 2023
  • Miscellaneous

    The initial label for me is a sports fan and I love watching various sports games in my leisure time. My favorite players are LeBron James, Lionel Messi, Roger Federer for basketball, football, and tennis respectively.

    Furthermore, I will also spend some time in watching popularization videos in different subjects to prevent me from being 'ignorant' in other fields.


    © Xuanbai Chen | Last updated: Mar 21, 2024 | Jon Barron's webpage template