yu_avatar.jpg

Yu Tian

Yu Tian is a postdoctoral research fellow at Harvard University. He received his Ph.D. in computer science at the Australian Institute of Machine Learning (AIML), University of Adelaide. He was also affiliated with South Australian Health and Medical Research Institute (SAHMRI) during his Ph.D. Previously, Dr. Tian obtained his bachelor’s degree in computer science with First Class Honours at the University of Adelaide.

Research

My main research interests are in the fields of computer vision and medical image analysis, in particular abnormality and rarity learning tasks, such as image/video anomaly detection for surveillance and industrial applications or early detection of diseases.

News

Nov, 2022 Invited to give a talk at USTC Suzhou Institute of Advanced Research about my recent works in anomaly/ood detection.
Sep, 2022 Invited to give a talk at GESA Research Workshop 2022 about Anomaly Detection in Medical Imaging.
Aug, 2022 I have finished my PhD and joined Harvard Ophthalmology AI Lab as a postdoctoral research fellow.
Jul, 2022 One paper on anomaly/ood segmentation for urban driving scenes accepted to ECCV 2022 - selected for oral presentation.
May, 2022 Invited to be the reviewer of IEEE Transactions on Image Processing.
May, 2022 Four papers are early accepted to MICCAI 2022.

Selected Publications

* denotes equal contribution; ^ denotes corresponding authorship.

  1. Pixel-wise Energy-biased Abstention Learning for Anomaly Segmentation on Complex Urban Driving Scenes
    Yu Tian*, Yuyuan Liu*, Guansong Pang, Fengbei Liu, Yuanhong Chen, and Gustavo Carneiro
    European Conference on Computer Vision (ECCV Oral) 2022
    PDF Code
  2. Contrastive Transformer-based Multiple Instance Learning for Weakly Supervised Polyp Frame Detection
    Yu Tian, Guansong Pang, Fengbei Liu, Yuyuan Liu, Chong Wang, Yuanhong Chen, Johan W Verjans, and Gustavo Carneiro
    International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI Early Accept) 2022
    PDF Code
  3. ACPL: Anti-curriculum Pseudo-labelling for Semi-supervised Medical Image Classification
    Fengbei Liu*, Yu Tian*, Yuanhong Chen, Yuyuan Liu, Vasileios Belagiannis, and Gustavo Carneiro
    IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2022
    PDF Code
  4. Deep One-Class Classification via Interpolated Gaussian Descriptor
    Yuanhong Chen*, Yu Tian*^, Guansong Pang, and Gustavo Carneiro
    In Thirty-Sixth AAAI Conference on Artificial Intelligence (AAAI Oral) 2022
    PDF Code
  5. Weakly-supervised Video Anomaly Detection with Robust Temporal Feature Magnitude Learning
    Yu Tian, Guansong Pang, Yuanhong Chen, Rajvinder Singh, Johan W Verjans, and Gustavo Carneiro
    In Proceedings of the IEEE/CVF international conference on computer vision (ICCV) 2021
    PDF Code
  6. Constrained Contrastive Distribution Learning for Unsupervised Anomaly Detection and Localisation in Medical Images
    Yu Tian, Guansong Pang, Fengbei Liu, Seon Ho Shin, Johan W Verjans, Rajvinder Singh, Gustavo Carneiro, and  others
    In International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2021
    PDF Code
  7. Few-shot anomaly detection for polyp frames from colonoscopy
    Yu Tian, Gabriel Maicas, Leonardo Zorron Cheng Tao Pu, Rajvinder Singh, Johan W Verjans, and Gustavo Carneiro
    In International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2020
    PDF Code