Name: Liu Li
Title: Associate Professor

Research Interests:
Machine Learning, Deep Learning, Multimodal Large Models

Education:

· Ph.D. in Engineering, East China Normal University, 2015

· Joint Ph.D. training at Concordia University, Canada (2013–2014)

Professional Experience:

· 09/2015–09/2017: University of Shanghai for Science and Technology

· 06/2016–09/2016: Concordia University

· 09/2017–Present: Nanchang University

Academic Service:

· Reviewer for multiple SCI journals, including IEEE Transactions on Image Processing and Pattern Recognition

· Member of the Document Image Analysis and Recognition Technical Committee, CSIG

Major Achievements:

1. Selected Research Projects Led:

o 01/2026–12/2029: Self-Supervised Style Representation Learning and Fine-Grained Disentanglement for the Digitization of Chinese Calligraphy, National Natural Science Foundation of China (Grant No. 62566237), Principal Investigator, ongoing

o 01/2017–12/2019: Near-Duplicate Text Image Matching Based on Keypoint Graph Representation, National Natural Science Foundation of China (Grant No. 61603256), Principal Investigator, completed

2. Selected Publications

[1] Luo M, Liu L^*, Lu Y, et al. Art style classification via self-supervised dual-teacher knowledge distillation [J]. Applied Soft Computing, 2025: 112694.

[2] Liu L, Xiong X, Wan M, et al. Chinese calligraphy character generation with component-level style learning and structure-aware guidance[J]. Applied Soft Computing, 2025: 113159.

[3] Zhang W, Ma H, Liu L*, et al. CalliNet: a triplet network for chinese calligraphy style classification[J]. International Journal on Document Analysis and Recognition (IJDAR), 2025: 1-16.

[4] Huang J, Liu L*, Lu Y, et al. Enhancing scene text script identification through multi-task self-supervised learning: J. Huang et al[J]. The Visual Computer, 2025: 1-16.

[5] Liu L, Lu Y, Suen C Y. End-to-end learning of representations for instance-level document image retrieval[J]. Applied Soft Computing, 2023: 110136.

[6] Cheng Wenyen, Zhou Yong, Tao Chengying, Liu L*, et al. Chinese calligraphy font and style classification based on a multi-loss fusion network. Journal of Image and Graphics, 2023, 28(8): 2370–2381.

[7] Peng F, Ma H, Liu L^*, et al. Adaptive feature fusion for scene text script identification[J]. Multimedia Tools and Applications, 2024: 1-23.

[8] Zhou Y, Ma H, Liu L^*, et al. Feature fusion and decomposition: exploring a new way for Chinese calligraphy style classification[J]. The Visual Computer, 2023: 1-12.

[9] Li Z, Liu L^*, Qiu T, et al. Modeling Cross-layer Interaction for Chinese Calligraphy Style Classification[C]// Proceedings of International Conference on Document Analysis and Recognition, 2023: 70-84.

[10] Xu Baixiang, Liu L*, Qiu Taorong. A triple-branch Siamese network for near-duplicate text image retrieval. Journal of Intelligent Systems, 2022, 17(3): 515–522.

[11] Liu L, Wang Z, Qiu T, et al. Document image classification: Progress over two decades[J]. Neurocomputing, 2021, 453: 223-240.

[12] Liu L, Qiu T, Lu Y, et al. Combination of spatially enhanced bag-of-visual-words model and genuine difference subspace for fake coin detection[J]. Expert Systems with Applications, 2020, 159: 113551.

[13] Liu L, Lu Y, Suen C Y. An image-based approach to detection of fake coins[J]. IEEE Transactions on Information Forensics and Security, 2017, 12(5): 1227-1239.

[14] Yang H, Liu L, Min W, et al. Driver yawning detection based on subtle facial action recognition[J]. IEEE Transactions on Multimedia, 2020, 23: 572-583. [15] Liu L, Lu Y, Suen C Y. Variable-length signature for near-duplicate image matching[J]. IEEE Transactions on Image Processing, 2015, 24(4): 1282-1296.

[16] Liu L, Cheng W, Qiu T, et al. Multi-loss siamese convolutional neural network for Chinese calligraphy style classification[C]//Proceedings of International Conference on Neural Information Processing, 2021: 425-432.

[17] Deng M, Ma H, Liu L^*, et al. ScriptNet：A two stream CNN for script identification in camera-based document images [C]//Proceedings of International Conference on Neural Information Processing, 2022.

[18] Liu L, Lu Y, Suen C Y. Near-duplicate document image matching: A graphical perspective[J]. Pattern Recognition, 2014, 47(4): 1653-1663.

[19] Liu L, Lu Y, Suen C Y. Novel global and local features for near-duplicate document image matching[C]// Proceedings of International Conference on Pattern Recognition, 2014: 4624-4629.

[20] Liu L, Lu Y, Suen C Y, et al. Modeling Local Word Spatial Configurations for Near Duplicate Document Image Retrieval[C]// Proceedings of International Conference on Document Analysis and Recognition, 2013: 235-239.

[21] Liu L, Lu Y, Suen C Y. Document image matching using probabilistic graphical models[C]//Proceedings of International Conference on Pattern Recognition, 2012: 637-640.

(3) Student Supervision

· Awarded the Outstanding Instructor title multiple times in the Lanqiao Cup Programming Competition.

· Recipient of the Outstanding Master’s Thesis Supervisor Award at Nanchang University (Class of 2025).

o Thesis title: Chinese Calligraphy Font Generation Based on Component-Level Style Learning and Structure-Aware Guidance.

· Recipient of the Outstanding Undergraduate Graduation Project (Thesis) Supervisor Award at Nanchang University (Class of 2025).

o Thesis title: Deep Learning–Based Classification of Chest X-ray Images.

· In 2025, two graduate students from the laboratory were awarded the Outstanding Graduate honor at Nanchang University.

· In 2024, one graduate student from the laboratory was awarded the Outstanding Graduate honor at Nanchang University.

· Supervised four graduate students who successfully obtained the Jiangxi Provincial Graduate Innovation Fund.

Email: liuli_033@163.com