Short Bio

Hi, here is Yanwu Xu, currently a 4th-year PhD student in Intellegent System at University of Pittsburgh, advised by Dr. Kayhan Batmanghelich and co-advised by Dr. Mingming Gong. My current research interests are Generative Adversarial Learning and Image Style Transfer. Before that, I obtained B.S. degree in Electrical Mechanical Engineering department at Central South University in 2017.

Blog

  • Congrats to our Image2Image translation model accpeted by CVPR2022. In this work, we proposed an adversarial spatial perturbation consistency for I2I task. More detials, please refer to our git repo.
    Unpaired image-to-image translation (I2I) is an ill-posed problem, as an infinite number of translation functions can map the source domain distribution to the target distribution. Therefore, much effort has been put into designing suitable constraints, e.g., cycle consistency (CycleGAN), geometry consistency (GCGAN), and contrastive learning-based constraints (CUTGAN), that help better pose the problem. However, these well-known constraints have limitations: (1) they are either too restrictive or too weak for specific I2I tasks; (2) these methods result in content distortion when there is a significant spatial variation between the source and target domains. This paper proposes a universal regularization technique called maximum spatial perturbation consistency (MSPC), which enforces a spatial perturbation function ($T$) and the translation operator ($G$) to be commutative (i.e., $T \circ G = G \circ T $). In addition, we introduce two adversarial training components for learning the spatial perturbation function. The first one lets $T$ compete with $G$ to achieve maximum perturbation. The second one lets $G$ and $T$ compete with discriminators to align the spatial variations caused by the change of object size, object distortion, background interruptions, etc. Our method outperforms the state-of-the-art methods on most I2I benchmarks. We also introduce a new benchmark, namely the front face to profile face dataset, to emphasize the underlying challenges of I2I for real-world applications. We finally perform ablation experiments to study the sensitivity of our method to the severity of spatial perturbation and its effectiveness for distribution alignment.

News

  • One paper accepted by MICCAI2022.

  • I will start my 2022 summer internship at Google.

  • One papers is accepted by CVPR 2022.

  • One papers is accepted by ACM 2021.

  • One papers is accepted by ICCV 2021.

  • Two papers are accepted by AAAI 2020.

  • One paper is accepted by NeurIPS 2019.

  • One paper is accepted by BraTS Challenge 2018.

  • One paper is accepted by ACCV 2018.