Boxuan Zhang

Boxuan Zhang

Master of Artificial Intelligence

Wuhan University

School of Computer Science

Biography

I am currently on M.Eng program in artificial intelligence at Sensing IntelliGence and MAchine learning(SIGMA) Lab in the School of Computer Science, Wuhan University, advised by Dr. Zengmao Wang and Prof. Bo Du. Before this I completed my B.Eng in Wuhan University of Computer Science and Technology. I am seeking 24fall Ph.D opportunity now.

My research interests include machine learning(e.g., self-supervised & semi-supervised learning & active learning, reliable machine learning like learning and inference with Out-of-distribution data) and computer vision(e.g., classification and object detection). You can email me at zhangboxuan1005 [at] gmail [dot] com.

Interests
  • Machine Learning
  • Computer Vision
Education
  • MEng in Artificial Intelligence, 2022 - 2024(Expected)

    Wuhan University, China

  • BEng in Computer Science and Technology, 2018 - 2022

    Wuhan University, China

Recent News

- [Jan. 2024] Our work SSOD-AT is accepted by GRSL.
- [Nov. 2023] Start working with Jianing Zhu in the TMLR Group@ HKBU on OOD detection.
- [Oct. 2023] Honored to receive 3rd Award @ TBM Machine Learning Competition.
- [Oct. 2023] Attend 2nd TBM Machine Learning Competition(held by CSRME) and present research work @ Shanghai.
- [Sep. 2023] Start my second-year research & learning journey in WHU.
- [Aug. 2023] Welcome to check the project page of my first work SSOD-AT, which has been submitted to GRSL.

Publications

(0001). Boosting Semi-Supervised Object Detection in Remote Sensing Images with Active Teaching. Accepted by Geoscience and Remote Sensing Letters(GRSL).

Cite DOI Code PDF

Projects

*
Machine Learning on TBMs Excavation
A study of rock mass accurate classification based on multi-algorithm cross multi-feature optimization selection and TBM parameter efficient prediction using low-dimensional inputs.
Machine Learning on TBMs Excavation
Self-Supervised Techniques for Intelligent Image Annotation
This study is aimed at ”smart city” and ”smart healthcare”, and is committed to achieving or even surpassing the performance of a large number of labeled images by using only image-level annotations or unlabeled images.
Self-Supervised Techniques for Intelligent Image Annotation

Contact

7 Pageviews
Jan. 03rd - Feb. 03rd