Ruixuan Wang

Associate Professor
Intelligence Science and System Lab (iSEE)
School of Data and Computer Science
Sun Yat-sen University, China

Email: wangruix5[at]mail[dot]sysu[dot]edu[dot]cn


My dream is to create intelligent machines having human-level capabilities, particularly the ability of perceiving and understanding the world through vision. With this long-term motivation, I obtained relevant Bachelor and Master degrees both from Xian Jiaotong University, and the PhD degree from National University of Singapore, followed by the participation of multiple interesting AI-relevant research programs mainly in University of Dundee, UK. In the past ten years, I witnessed and joined the wave of the shift in AI approaches from the old feature engineering to the new Deep Learning. However, Deep Learning itself cannot achieve the artificial general intelligence (AGI). Creative solutions and novel learning mechanisms need to be developed to solve challenging fundamental problems. Often, these challenging problems are derived from practical applications such as autonomous driving, surveillance, and healthcare. Our research group uses healthcare applications (particularly relevant to medical image analysis) as the engine, driving the exploration and development of novel AI techniques and solutions. If you have a similar AI dream, either planning to study basic AI knowledge and techniques, or looking for a laboratory to continue your AI research, please feel free to drop me an email - we are always looking for highly self-motivated undergraduates, postgraduates, postdocs, and research professors to join us! [Prospective students click here]

Selected Publications

  1. Rong Zhang, Shuhan Tan, Ruixuan Wang, Siyamalan Manivannan, Jingjing Chen, Haotian Lin, Weishi Zheng.
    Biomarker Localization by Combining CNN Classifier and Generative Adversarial Network,
    International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2019. [Paper]
  2. Feifei Xue, Jin Peng, Ruixuan Wang, Qiong Zhang, Weishi Zheng.
    Improving Robustness of Medical Image Diagnosis with Denoising Convolutional Neural Networks,
    International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2019. [Paper]
  3. Zhiying Cui, Longshi Wu, Ruixuan Wang, Weishi Zheng.
    Ensemble Transductive Learning for Skin Lesion Segmentation,
    Chinese Conference on Pattern Recognition and Computer Vision (PRCV), 2019. [Paper]
  4. Jiaxin Zhuang, Jiabin Cai, Ruixuan Wang, Jianguo Zhang, Weishi Zheng.
    CARE: Class Attention to Regions of Lesion for Classification on Imbalanced Data,
    International Conference on Medical Imaging with Deep Learning (MIDL), 2019. [Paper]
  5. Hongwei Li, Gongfa Jiang, Jianguo Zhang, Ruixuan Wang, Zhaolei Wang, Wei-shi Zheng, Bjoern Menze.
    Fully convolutional network ensembles for white matter hyperintensities segmentation in MR images,
    NeuroImage, 2018. [Paper]
  6. Gareth J. McKay, Euan N. Paterson, Alexander P. Maxwell, Christopher C. Cardwell, Ruixuan Wang, Stephen Hogg, Thomas J. MacGillivray, Emanuele Trucco, Alexander S. Doney.
    Retinal microvascular parameters are not associated with reduced renal function in a study of individuals with type 2 diabetes,
    Scientific Report, 8:3931, 2018. [Paper]
  7. Haocheng Shen, Ruixuan Wang, Jianguo Zhang, Stephen McKenna.
    Boundary-aware Fully Convolutional Network for Brain Tumor Segmentation,
    International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2017. [Paper]
  8. Haocheng Shen, Ruixuan Wang, Jianguo Zhang, Stephen McKenna.
    Multi-task Fully Convolutional Network for Brain Tumour Segmentation,
    Medical Image Understanding and Analysis (MIUA), 2017. [Paper]
  9. Siyamalan Manivannan, Wenqi Li, Shazia Akbar, Ruixuan Wang, Jianguo Zhang, Stephen J. McKenna.
    An Automated Pattern Recognition System for Classifying Indirect Immunofluorescence Images of HEp-2 Cells and Specimens,
    Pattern Recognition, 51, 12-26, 2016. [Paper]
  10. Siyamalan Manivannan, Ruixuan Wang, Emanuele Trucco.
    Hierarchical Mix-Pooling and its Applications to Biomedical Image Classification,
    International Symposium on Biomedical Imaging (ISBI), 2016. [Paper]
  11. Ruixuan Wang, Markus Pakleppa, Emanuele Trucco.
    Low Rank Prior in Single Patches for Non-pointwise Pulse Noise Removal,
    IEEE Transactions on Image Processing, 25(4), 1485-1496, 2015. [Paper]
  12. Siyamalan Manivannan, Ruixuan Wang, Emanuele Trucco.
    Inter-cluster Features for Medical Image Classification,
    International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2014. [Paper]
  13. Ruixuan Wang, Emanuele Trucco.
    Single-patch Low-rank Prior for Non-pointwise Impulse Noise Removal,
    IEEE International Conference on Computer Vision (ICCV), 2013. [Paper]
  14. Siyamalan Manivannan, Ruixuan Wang, Emanuele Trucco, Adrian Hood.
    Automatic Normal-abnormal Video Frame Classification for Colonoscopy,
    International Symposium on Biomedical Imaging (ISBI), 2013. [Paper]
  15. Wee Kheng Leow, Ruixuan Wang, Hong Wai Leong.
    3D-2D Spatiotemporal Registration for Sports Motion Analysis.
    Machine Vision and Applications, 23(6), 1177-1194, 2012. [Paper]
  16. Ruixuan Wang, Himanshu Pokhariya, Stephen J. McKenna, John Lucocq.
    Recognition of Immuno Gold Markers in Electron Micrographs.
    Journal of Structural Biology, 176(2), 151-158, 2011. [Paper]
  17. Ruixuan Wang, Stephen J. McKenna, Junwei Han.
    Visualizing Image Collections using High-Entropy Layout Distributions,
    IEEE Transaction on Multimedia, 12(18), 803-813, 2010. [Paper]
  18. Ruixuan Wang, Stephen J. McKenna.
    Gaussian Process Learning from Order Relationships using Expectation Propagation,
    International Conference on Pattern Recognition (ICPR), 2010. [Paper]
  19. Junwei Han, Stephen J. McKenna, Ruixuan Wang.
    Learning Query-Dependent Distance Metrics for Interactive Image Retrieval,
    International Conference on Vision Systems (ICVS), 2009. [Paper]
  20. Ruixuan Wang, Stephen J. McKenna, Junwei Han.
    High-Entropy Layouts for Content-based Browsing and Retrieval,
    ACM International Conference on Image and Video Retrieval (CIVR), 2009. [Paper]
  21. Junwei Han, Stephen J. McKenna, Ruixuan Wang.
    Regular Texture Analysis as Statistical Model Selection,
    European Conference on Computer Vision (ECCV), 2008. [Paper]
  22. Ruixuan Wang, Wee Kheng Leow, Hong Wai Leong.
    3D-2D Spatiotemporal Registration for Sports Motion Analysis,
    IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2008. [Paper]
  23. Patents:
  24. Stephen J. McKenna, Ruixuan Wang, Annette Ward.
    System and Method for Arranging Items for Display,
    China Patent CN102395963, 2014; US Patent 9229939, 2016.
  25. Nanning Zheng, Ruixuan Wang, Yong Wu, Guanglie Zhang, Weipu Xu.
    An Internally Parallel Method for VLSI Implementation of 2D Discrete Wavelet Transform,
    China patent CN1215553, 2005.
Course: Deep Learning

Claim: some slides were copied or adapted from other online courses, and all rights of these slides belong to their respective owners! Notice: slides may be updated over months, based on your comments and new interesting papers.

  1. Week 1 (2019.02.28): Introduction [slides]
    Applications of deep learning, general procedure of problem solving, basic neural network structure and units.
  2. Week 2 (2019.03.07): Optimization [slides]
    Backpropagation, optimization methods, hyper-parameter tuning, deep learning frameworks.
  3. Week 3 (2019.03.14): Training issues [slides]
    Gradient exploding and vanishing, weight initialization, batch normalization, methods to reduce overfitting.
  4. Week 4 (2019.03.21): CNN-1 [slides]
    Convolution, kernel, channel, stride, pooling, end-to-end, CNN architecture, AlexNet, VggNet, ResNet.
  5. Week 5 (2019.03.28): CNN-2 [slides]
    DenseNet, wide ResNet, ResNeXt, SENet, PNASNet, FaceNet, style transfer, text classification.
  6. Week 6 (2019.04.04): Segmentation [slides]
    FCN, SegNet, transpose convolution, U-net, PSPNet, dilated convolution, DeepLab v1 - v3+, mixed-scale dilated FCN .
  7. Week 7 (2019.04.11): Detection [slides]
    2-stage models: R-CNN, fast R-CNN, faster R-CNN, FPN, mask R-CNN; 1-stage models: YOLO, SSD, RetinaNet, FCOS; DetNet.
  8. Week 8 (2019.04.18): GAN-1 [slides]
    Generator and Discriminator, DCGAN, LSGAN, training issues, Wasserstein distance, Wasserstein GAN, WGAN-GP.
  9. Week 9 (2019.04.25): GAN-2 [slides]
    Conditional GAN; CGAN applications: StackGAN, image-image translation; CycleGAN, MUNIT, CAN, SRGAN, progressive GAN.
  10. Week 10 (2019.04.28): RNN-1 [slides]
    Word2vec, language models, RNN basics, RNN training, LSTM, GRU, Bidirectional RNN, deep RNN.
  11. Week 11 (2019.05.09): RNN-2 [slides]
    Encoder-decoder for machine translation, beam search, attention mechanism, poetry generation, dialogue models.
  12. Week 12 (2019.05.16): Memory networks [slides]
    Question answering, memory network (MN), dynamic MN, key-value MN, recurrent entity network, neural Turing machine.
  13. Week 13 (2019.05.23): Vision to language [slides]
    Image caption, show and tell, attribute CNN, show_attend_and_tell, adaptive attention, spatial attention, neural baby talk.
  14. Week 14 (2019.05.30): Interpretability-1 [slides]
    Visualization of filters, deconvolution, network dissection, net2vec, activation maximization, gradient-based, perturbation based.
  15. Week 15 (2019.06.06): Interpretability-2 [slides]
    Feature map activation (CAM, Grd-CAM), basis decomposition, GAN interpretation, (simple) theoretical analysis .
  16. Week 16 (2019.06.13): Robustness [slides]
    Adversarial examples; attacks: FGSM and variants, C-W method; defense: training with adv. examples, distillation, MagNet, defense GAN.
  17. Week 17 (2019.06.20): Efficiency [slides]
    Factorization, knowledge transfer, pruning, quantization (XNOR-net, DoReFa-net), novel designs (SqueezeNet, MobileNet, ShuffleNet).
  18. Week 18 (2019.06.27): Trends [slides]
    Few-shot/meta learning: matching network, MAML; life-long learning: EWC and variants, deep generative replay; learning from X... .

  1. Third place (out of 77 teams), MICCAI 2018 grand challenge "ISIC 2018: Skin Lesion Analysis Towards Melanoma Detection", Task 3 "Lesion Diagnosis", 2018. Note: top 2 teams used additional public dataset to train models.
  2. Winner (1st place), MICCAI 2017 grand challenge "WMH Segmentation Challenge", 2017.
  3. Recipient of "Hundred-Talent Program" (˼ƻ), Sun Yat-sen University, 2017.
  4. Winner (1st place), Cell Level Classification task in ICPR contest "Performance Evaluation of Indirect Immunofluorescence Image Analysis Systems", 2014.
  5. Winner (1st place), Specimen Level Classification task in ICPR contest "Performance Evaluation of Indirect Immunofluorescence Image Analysis Systems", 2014.
  6. Second prize, MICCAI 2014 grand challenge "Brain Tumor Digital Pathology Challenge", Sub-challenge 2: Segmentation, 2014.
  7. Discipline Hopping Award, UK MRC/BBSRC/EPSRC Councils, 2010-2011.

  1. Big Data Mining and Integrated Platform Development for Precision Medcine, Guangdong Key Research and Development Program, 8,000,000 RMB, Co-investigator, 2019-2021.
  2. Clinical Data Standardization and Applications for the Study of Major Chronic Diseases, National Key Research and Development Program, 8,040,000 RMB, Co-investigator, 2018-2020.
  3. Deep Learning based Anomaly Detection with Large Healthy Data, Guangzhou Science and Technology Program, 200,000 RMB, Principle investigator, 2019-2021.
  4. Medical Image Analysis with Deep Learning, "Hundred-Talent Program" funding, Sun Yat-sen University, 450,000+ RMB, Principle investigator, 2018-2019.
  5. Discovering Retinal Biomarkers for Diabetes with Vampire, UK Wellcome Trust, 160,000, Lead researcher, 2016-2017.
  6. Colonic Disease Investigation by Robot Hydro-colonoscopy (CODIR), EU European Research Council Advanced Grant, Euros 3,000,000, Lead researcher, 2011-2014.
  7. Microscopic Image Analysis for Cell Biology, MRC/ BBSRC/EPSRC Discipline Hopping Award, 116,000, Lead researcher, 2010-2011.
  8. FABRIC: Fashion and Apparel Browsing for Inspirational Content, Technology Strategy Board grant, 924,000, Lead researcher, 2007-2010.