Professor Niu Sijie
(牛四杰)Professor & Doctoral Supervisor
Professor, PhD, Doctoral Supervisor. Leader of Shandong Province Youth Innovation Team, Haiyou Industrial Leader, and ACM Jinan Rising Star Award Winner. PhD graduate from Nanjing University of Science and Technology (2016) in Pattern Recognition and Intelligent Systems. Visiting scholar at Stanford University (2014) funded by China Scholarship Council. Postdoctoral researcher at UNC Chapel Hill (Dec 2019 - Jan 2021) with Prof. Dinggang Shen. Has led 7 research projects including NSFC Youth Project, Shandong Natural Science Foundation, and China Postdoctoral Science Foundation. Published 59 papers in IEEE series, Pattern Recognition, Information Fusion, Ophthalmology, with 18 SCI-indexed papers as first/corresponding author. One paper was selected as top 1% ESI Highly Cited Paper for five consecutive years (2017-2021). Applied for 9 national invention patents, published 1 monograph, and 4 software copyrights. Received ACM China Council Jinan Outstanding Doctoral Dissertation Award and NJUST Outstanding Doctoral Dissertation Award.
Research Metrics
Research Interests
Few-shot medical anomaly detection through centroid consultation back and test-time self-calibration
Towards reliable medical unsupervised anomaly detection: A benchmark and dataset for PET/CT with cross-modality knowledge distillation network
Enhancing Diffusion Models Towards Anomaly‐Aware Reconstructions for Medical Image Anomaly Detection
FGCLIP-based Augmented Language-driven Contrastive Clustering Network for Fine-grained Image Clustering
Semantic Augmentation Variational Autoencoder for Unsupervised Anomaly Detection in Retinal OCT Images
Advancements in Spiking Neural Networks for Image Recognition: A Review of Research Progress
Active Projects
CNN Model Compression via Sparse Representation
Research on compressing convolutional neural networks using sparse self-representation and particle swarm optimization techniques. Achieved significant model size reduction while maintaining accuracy for real-world deployment.
Medical Image Segmentation for Retinal Disease Detection
Development of advanced segmentation algorithms for retinal layer analysis and disease detection in OCT images. Focus on central serous chorioretinopathy and other retinal pathologies using deep learning and P systems.
Hyperspectral Image Analysis for Remote Sensing
Research on sparse subspace clustering and high-order methods for hyperspectral image classification and interpretation in remote sensing applications.
Face Anti-Spoofing and Liveness Detection
Development of intelligent face liveness detection platform using multi-feature fusion and deep learning for security applications.
Retinal Image Super-Resolution via Dictionary Learning
Research on dictionary learning methods for enhancing retinal image resolution for improved disease diagnosis and analysis.