Xiaofan Zhang

EDUCATION RESEARCH INTEREST RESEARCH EXPERIENCE PUBLICATION SKILLS

EDUCATION

UNC Charlotte 8/2013 - Present
Ph.D. in Computer Science

Beihang University(BUAA) 9/2009 - 7/2013
B.E. in Automation Science and Electrical Engineering

RESEARCH INTEREST

Computer vision, medical image analysis, machine learning, fine-grained image recognition

5/2016 - 8/2016 Research Intern
Media Analytics, NEC Laboratories America
Mentor: Dr. Kihyuk Sohn
Propose to learn robust features that are invariant to the identification-irrelevant attributes (e.g., expressions, poses) base on triplet network.

Fine-Grained Image Recognition

5/2015 - 8/2015 Research Intern
Media Analytics, NEC Laboratories America
Mentor: Dr. Feng Zhou, Dr. Yuanqing Lin
1. Proposed a multi-task learning framework to embed structured labels and learn fine-grained feature representations effectively.
2. Combined the Spatial Transformer Network (STN) and LSTM for accurate parts detection to boost the performance.

Large-Scale Image Retrieval for Medical Image Analysis

8/2013 - Present Ph.D. Student
VIA Lab, UNC Charlotte
Advisor: Dr. Shaoting Zhang (Assistant Professor)
Proposed hashing-based large-scale image retrieval methods to analyze histopathological images, achieving 88.1% classification accuracy within 20 milliseconds, within large medical image databases.

Urine Sediment Analysis

7/2012 - 1/2013 Research Intern
Imaging and Data Processing for Emerging Markets Group, Siemens Corporate Technology, Beijing, China
Mentor: Dr. Tian Shen (Research Scientist)
Developed a framework to classify particles in urine sediment based on texture features and random forest. Two pending patents.

Image Retrieval, Object recognition

9/2011 - 7/2013 Research Student
Intelligent Computing and Machine Learning Lab, Beihang University
Advisor: Dr. Zengchang Qin (Associate Professor)
Developed and implemented over 20 classical features/methods, including SIFT, GIST, HOG and its variants. Proposed an effective feature for object recognition and image understanding, based on saliency detection and image segmentation.

PUBLICATION

Embedding Label Structures for Fine-grained Feature Representation. [PDF]
Xiaofan Zhang, Feng Zhou, Yuanqing Lin, Shaoting Zhang.
Computer Vision and Pattern Recognition (CVPR), 2016.

Fusing Heterogeneous Features from Stacked Sparse Autoencoder for Histopathological Image Analysis. [PDF]
Xiaofan Zhang, Hang Dou, Tao Ju, Jun Xu, Shaoting Zhang.
IEEE Journal of Biomedical and Health Informatics (JBHI), Volume 20, Issue 5, Pages 1377-1383, 2016.

High-Throughput Histopathological Image Analysis via Robust Cell Segmentation and Hashing. [PDF]
Xiaofan Zhang, Fuyong Xing, Hai Su, Lin Yang, Shaoting Zhang.
Medical Image Analysis (MedIA), Volume 26, Issue 1, Pages 306–315, 2015.

Fine-Grained Histopathological Image Analysis via Robust Segmentation and Large-Scale Retrieval. [PDF]
Xiaofan Zhang, Hai Su, Lin Yang, Shaoting Zhang.
Computer Vision and Pattern Recognition (CVPR), 2015.

Weighted Hashing with Multiple Cues for Cell-Level Analysis of Histopathological Images. [PDF]
Xiaofan Zhang, Hai Su, Lin Yang, Shaoting Zhang.
The 24th biennial international conference on Information Processing in Medical Imaging (IPMI)., 2015.

Fusing Heterogeneous Features for the Image-Guided Diagnosis of Intraductal Breast Lesions. [PDF]
Xiaofan Zhang, Hang Dou, Tao Ju, Shaoting Zhang.
International Symposium on Biomedical Imaging (ISBI), 2015. oral presentation, ~130 / 700 = ~18%
Best Student Paper Awards Finalists(18 out of ~700 submissions)

Towards Large-Scale Histopathological Image Analysis: Hashing-Based Image Retrieval. [PDF]
Xiaofan Zhang, Wei Liu, Murat Dundar, Badve Sunil, Shaoting Zhang.
IEEE Transactions on Medical Imaging (TMI), 2015.

Mining Histopathological Images via Composite Hashing and Online Learning. [PDF]
Xiaofan Zhang, Lin Yang, Wei Liu, Hai Su, Shaoting Zhang.
International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), 2014.
Early acceptance rate, ~10%. Student travel award

Mining Histopathological Images via Hashing-Based Scalable Image Retrieval. [PDF]
Xiaofan Zhang, Wei Liu, Shaoting Zhang.
The IEEE International Symposium on Biomedical Imaging (ISBI), 2014. oral presentation, ~120 / 600 = ~20%
Best Paper Travel Award (3 awardees out of ~600 submissions)

SKILLS

Proficient in C/C++, Matlab, Python;
Proficient in Linux;
Proficient in deep learning.