Please visit my New WebPage . This website will no longer be updated.

I am Wei Xiong. I am a first year Ph.D. candidate at Department of Computer Science, University of Rochester. My advisor is Prof. Jiebo Luo. My research interests include Computer Vision and Machine Learning. For more details, see my [CV] here.



Learn about what I do

Projects

Deep Unsupervised Network for Visual Rocognition

  • We develop the multi-layer unsupervised neural networks named Stacked Convolutional Denoising Auto-Encoders for object recognition tasks.
  • Our model is composed of layers of Auto-Encoders and the parameters are updated by greedy layer-wise training.

Structured Decorrelated Constraint for Model Regularization

  • We propose a regularization method to aleviate the overfitting problem in CNNs.
  • Our Structured Decorrelation Constraint can be applied to both the convolutional feature maps and the fully connected layers. It can be optimized along with the classification loss.

Rich and Robust Feature Pooling in Unsupervised Framework

  • We propose a novel pooling method to generate better statistics from the feature maps leant by the unsupervised frameworks.
  • Our method can extract rich and robust features from the feature maps, outperforming the widely used Max pooling, Average pooling and Stochastic pooling.

Publications

Journals

Pattern Recognition

Wei Xiong, Lefei Zhang, Bo Du, Dacheng Tao. Combining local and global: Rich and robust feature pooling for visual recognition. Pattern Recognition 62: 225-235 (2017)

Trans. Cybernetics

Bo Du, Wei Xiong, Jia Wu, Lefei Zhang, Liangpei Zhang, Dacheng Tao: Stacked Convolutional Denoising Auto-Encoders for Feature Representation. IEEE Trans. Cybernetics 47(4): 1017-1027 (2017)

Conferences

CVPR'18

Wei Xiong, Wenhan Luo, Lin Ma, Wei Liu, and Jiebo Luo. Learning to generate time-lapse videos using multi-stage dynamic generative adversarial networks. Computer Vision and Pattern Recognition (CVPR), 2018. [PDF]

MMM'17

Fengling Mao, Wei Xiong, Bo Du, Lefei Zhang: Stochastic Decorrelation Constraint Regularized Auto-Encoder for Visual Recognition. MMM (2) 2017: 368-380.

ICDM'16

Wei Xiong, Bo Du, Lefei Zhang, Ruimin Hu, Dacheng Tao: Regularizing Deep Convolutional Neural Networks with a Structured Decorrelation Constraint. ICDM 2016: 519-528.

IJCNN'16

Wei Xiong, Bo Du, Lefei Zhang, Liangpei Zhang Dacheng Tao: Denoising auto-encoders toward robust unsupervised feature representation. IJCNN 2016: 4721-4728.

ICDM'15

Wei Xiong, Bo Du, Lefei Zhang, Ruimin Hu, Wei Bian, Jialie Shen, Dacheng Tao: R2FP: Rich and Robust Feature Pooling for Mining Visual Data. ICDM 2015: 469-478.

Courses

CSC 400 Program Seminar

ITRG Mini-Proposal [PDF]

CSC 440 Data Mining

Project: to be updated.

CSC 456 Operating Systems

Project: to be updated.

Contact

Email: wei.xiong@rochester.edu OR wxiongwhu@gmail.com

Office: Room 3602, Wegmans Hall, University of Rochester.

Please don't hesitate to contact me if you have any questions.