I am an Applied Scientist at Amazon AWS AI, Vancouver, Canada. My current research interest is the interaction between Computer Vision and Large Language Model.

Previously, I was a Research Associate at ETS Montreal, from November 2020 to March 2023, Canada. I worked at the Laboratoire d’Imagerie, de Vision et d’Intelligence Artificielle (LIVIA), under the supervision of Prof. Ismail Ben Ayed and Prof. Jose Dolz. Before that, I obtained some advanced industrial experiences (R&D for real products), as I worked for several companies (Coohom, Alibaba, etc.) from 2015 to 2020. I got my Ph.D at Institue of Automatoin, Chinese Academy of Sciences in 2015, where my supervisors were Prof. Hanqing Lu and Prof. Jing Liu. I received my BSc from Zhejiang University in 2010.

Besides tech stuff, I have passion in sports, reading and travel.

Interests
  • Computer Vision
  • Deep Learning
  • Large Language Model
Education

  • PhD in Computer Science, 2015

    Institute of Automation, Chinese Academy of Sciences

  • BSc in Information Science, 2010

    Zhejiang University

Recent Posts

🔥 One paper regarding model calibration is accepted at CVPR 2023
One paper regarding segmentation with mixed supervision is accpeted at MedIA
Our work on calibrating deep neural networks is accepted at CVPR 2022
Our results on MICCAI2021 GAMMA contest : Top 1 in two tasks

Featured Publications

Class Adaptive Network Calibration. CVPR, 2023.
Bingyuan Liu, Jérôme Rony, Adrian Galdran, Jose Dolz, Ismail Ben Ayed
arxiv code
Class Adaptive Network Calibration
Segmentation with mixed supervision: Confidence maximization helps knowledge distillation. MedIA, 2022.
Bingyuan Liu, Christian Desrosiers, Ismail Ben Ayed, Jose Dolz
DOI arxiv code
Segmentation with mixed supervision: Confidence maximization helps knowledge distillation
The Devil is in the Margin: Margin-based Label Smoothing for Network Calibration. CVPR, 2022.
Bingyuan Liu, Ismail Ben Ayed, Adrian Galdran, Jose Dolz
project arxiv code poster
The Devil is in the Margin: Margin-based Label Smoothing for Network Calibration
The hidden label-marginal biases of segmentation losses. arXiv, 2021.
Bingyuan Liu, Jose Dolz, Adrian Galdran, Riadh Kobbi, Ismail Ben Ayed
arxiv code
The hidden label-marginal biases of segmentation losses
Furnishing Your Room by What You See: An End-to-End Furniture Set Retrieval Framework with Rich Annotated Benchmark Dataset. arXiv, 2019.
Bingyuan Liu, Jiantao Zhang, Xiaoting Zhang, Wei Zhang, Chuanhui Yu, Yuan Zhou
arxiv project
Furnishing Your Room by What You See: An End-to-End Furniture Set Retrieval Framework with Rich Annotated Benchmark Dataset
See all publications

Popular Topics

calibration constrastive learning deep learning glaucoma grading image classification image retrieval image segmentation knowledge distillation macular fovea localization medical image segmentation multi-modal learning object detection optic disc and cup segmentation scene recognition sparse coding unsupervised learning

Contact