I'm a PhD candidate in Department of Electrical & Computer Engineering, University of Virginia. I work in the Imaging and Data Science Lab, supervised by Professor Gustavo K. Rohde. My research interests are in generative modeling, adversarial machine learning, and applying transport-based methods to signal processing and machine learning problems.

Contact: xy4cm@virginia.edu

Github: https://github.com/xuwangyin


Google Scholar

  • GAT: Generative Adversarial Training for Adversarial Example Detection and Robust Classification
    [code] [arXiv] [OpenReview] [Comprehensive Evaluation (Section 16)]
    Xuwang Yin, Soheil Kolouri, Gustavo K. Rohde
    International Conference on Learning Representations (ICLR 2020)

  • Radon Cumulative Distribution Transform Subspace Modeling for Image Classification [code] [arXiv]
    Mohammad Shifat-E-Rabbi, Xuwang Yin, Abu Hasnat Mohammad Rubaiyat, Shiying Li, Soheil Kolouri, Akram Aldroubi, Jonathan M. Nichols, Gustavo K. Rohde

  • Neural Networks, Hypersurfaces, and Radon Transforms [IEEEXplore Link][Code]
    Soheil Kolouri, Xuwang Yin, Gustavo K. Rohde
    IEEE Signal Processing Magazine

  • Cell image classification: a comparative overview [Link]
    M Shifat-E-Rabbi, Xuwang Yin, Cailey Elizabeth Fitzgerald, Gustavo K. Rohde.
    Cytometry Part A, the Journal of Quantitative Cell Science

  • Obj2Text: Generating Visually Descriptive Language from Object Layouts [link]
    Xuwang Yin, Vicente Ordonez
    Empirical Methods in Natural Language Processing (EMNLP 2017)

  • Robust Text Detection in Natural Scene Images [link]
    Xu-Cheng Yin, Xuwang Yin, Kaizhu Huang, Hong-Wei Hao
    IEEE transactions on pattern analysis and machine intelligence (TPAMI 2013)

  • Effective Text Localization in Natural Scene Images With MSER, Geometry-Based Grouping and AdaBoost [link]
    Xuwang Yin, Xu-Cheng Yin, Hong-Wei Hao, Khalid Iqbal
    International Conference on Pattern Recognition (ICPR2012)


  • PyTransKit [Github]
    Python Library for Signal/Image Data Analysis with Transport Methods.

  • Cell Image Classification Toolbox [Github]
    A toolbox for training cell images classification models. Supported models include neural networks (MLP, customized CNN, VGG, InceptionV3, ResNet, etc.), and statistical ML models (SVM, logitic regression, random forest, etc.). Support training with cross-validation, data augmentation and fine-tuning.


  • ICDAR2013 robust reading competition winner, International Association for Pattern Recognition, 2013