Ying, Fangli (应方立)
 |
Senior lecturer, MSc Supervisor, PhD Co-supervisor， |
About me
Dr. Fangli Ying (Github profile：fangli-ying.github.io) is a senior lecturer in the Department of Computer Science at East China University of Science and Technology (ECUST). He undertakes the supervision of both Master's and PhD students, serving as the co-supervisor for three international PhD students and the supervisor for eleven MSc students. Additionally, he holds the position of a visiting professor at the International College of Digital Innovation, Chiang Mai University (CMU), Thailand. His research focuses on the development of industrial applications of artificial intelligence, leveraging multidisciplinary approaches to address complex industrial challenges. He actively collaborates with multiple labs at ECUST, including the State Key Laboratory of Bioreactor Engineering, the Department of Finance, and the National Engineering Laboratory for Big Data Distribution and Exchange Technologies. His academic contributions are evidenced by numerous publications in prestigious journals and conferences, such as IEEE transactions, Applied Intelligence, Neural Computing and Applications, and ICME. Dr. Ying previously received the Best Runner-up Award from ACM SIGSPATIAL GIS conference and the ESRI European scholar Award. Dr. Ying has previously been honored with the Best Runner-up Award at the ACM SIGSPATIAL GIS conference and the ESRI European Scholar Award. His current research interests span across Computer Vision, Reinforcement Learning for Portfolio Management, and Bioinformatics.
Research
My research interests include:
Computational aesthetics and applications
Few-shot Generative Models
Deep Reinforcement for Portfolio Management
Deep Generative Models for Protein Design
Current work
-
Few-shot Image Generation
As a type of effective deep learning models, the Generative Adversarial Networks (GANs) are capable of synthesizing new realistic images and estimate the potential distribution of the samples utilizing adversarial learning. Nevertheless, the conventional GANs require a large amount of training data samples for producing plausible results. Inspired by the capacity of humans to quickly learn new concepts from a small number of examples, Fast Adaptive Meta-Learning (FAML) based GAN and encoder network is proposed in this study for few-shot image generation. This model demonstrates its capability to generate new realistic images from previously-unseen target classes with only a small number of examples required. By adopting two latent vectors, an enhanced algorithm is proposed to prevent mode collapse. Our model is able to improve few-shot image generation with the lowest FID score and highest IS on MNIST, Omniglot, VGG-Faces, and miniImageNet dataset. Moreover, it has been demonstrated the relativistic discriminator is effective in improving the overall visual quality of few-shots generation and is more than 10 times faster to achieve model convergence with as few as one-fourth of parameters required.
-
Reinforcement Learning for Portfolio Management
Although Deep Reinforcement Learning (DRL) models have made remarkable achievements in the financial trading, it is surprising that most of the literature ignores the possible risk of rare occurrences of catastrophic events and the effect of the worst-case scenarios on trading decisions. In this paper, we first develop a novel deep RL algorithm to control the risk of portfolio investment, while maximizing the α percentile expectation based on the distribution of future returns. A real world data set is used to validate the performance of our proposed models. The experimental results show that our proposed models outperform the market.
-
Video Summarization
With the exponential growth of video data, video summarization techniques are urgently needed for reducing people’s efforts in the videos' content exploration by generating succinct but informative summaries from original lengthy videos. Though supervised video summarization approaches have demonstrated the state-of-the-art performance, unsupervised methods are still highly demanded due to resourcefully expensive human annotations and the subjectiveness of video summarization tasks. In this paper, a novel unsupervised-based Deep Self-attention Recurrent summarization network with Reinforcement Learning (DSR-RL) for video summarization is proposed. The model can learn the input video sequence and suggest the key-shot summary without additional human annotations by integrating self-attention, BRNN, and reinforcement learning mechanisms. The DSR-RL improves not only importance score through the attention map vector of self-attention network but also the diversity of summaries via the reward function of reinforcement learning. Our method outperforms the state-of-the-art unsupervised video summarization methods on both SumMe and TVSum datasets.
-
Image Quality Assessment
Automatic facial beauty assessment has recently attracted a growing interest and achieved impressive results. However, despite the obvious subjectivity of beauty perception, most studies are addressed to predict generic or universal beauty and only few works investigate an individual’s preferences in facial attractiveness. Unlike universal beauty assessment, an effective personalized method is required to produce a reasonable accuracy on a small amount of training images as the number of annotated samples from an individual is limited in real-world applications. In this work, a novel personalized facial beauty assessment approach based on meta-learning is introduced. First of all, beauty preferences shared by an extensive number of individuals are learnt during meta-training. Then, the model is adapted to a new individual with a few rated image samples in the meta-testing phase. The experiments are conducted on a facial beauty dataset that includes faces of various ethnic, gender, age groups and rated by hundreds of volunteers with different social and cultural backgrounds. The results demonstrate that the proposed method is capable of effectively learning personal beauty preferences from a limited number of annotated images and outperforms the facial beauty prediction state-of-the-art on quantitative comparisons.
Accepted and Under Reviews
Fangli Ying*, Yunzhe Li, Yadan Yang, Aniwat Phaphuangwittayakul, Riyad Dhuny. "Enhancing Single-View 3D Clothed Human Reconstruction with Hybrid Prior Integration" COMPUTER GRAPHICS INTERNATIONAL (CGI) 2025 (Accepted)
Fangli Ying*, Ziyue Luo, Aniwat Phaphuangwittayakul and Yunze Li, "Enhancing Multimodal Video Summarization via Temporal and Semantic Alignment" The International Conference on Neural Information Processing(ICONIP) 2024 (Accepted)
Fangli Ying*, Ru Fan and Aniwat Phaphuangwittayaku, "Optimizing Risk Adaptation in Financial Trading with Multi-Scale LSTM Autoencoder and Predictive Transformer" Expert Systems and Applications (Under Review)
Fangli Ying*, Zhihong Zhang, Liting Zhou, Cathal Gurrin, Jinhai Wang, "Identity-Preserving Facial Aesthetic Enhancement via Hierarchical Prompt Learning and Pivotal Tuning" ACM MM 2025 (Under Review) [paper link in OpenReview]
Fangli Ying*, Zhihong Zhang, Aniwat Phaphuangwittayakul, Riyad Dhuny, " Identity-Preserving Facial Aesthetic Enhancement via Fine-Grained Attribute Prompting and Generator Tuning" ICCV 2025 (Under Review)[paper link in OpenReview]
Fangli Ying*, Zilong Li, Chaoqian Ouyang, Aniwat Phaphuangwittayakul ,Riyad Dhuny, Wilten Go. "De novo Highly Bioactive Multifunctional Antimicrobial Peptide Design with a Hybrid Generative Framework" International Journal of Molecular Sciences 2025 (Under Review)
Recent publications
- Jiang, Jieru, Fangli Ying*, and Riyad Dhuny. "Unveiling Technological Evolution with a Patent-Based Dynamic Topic Modeling Framework: A Case Study of Advanced 6G Technologies." Applied Sciences 15, no. 7 (2025): 3783.
- Dhuny, Riyad, and Fangli Ying. "Enhancing Education Accessibility: Portable Microservers for Computer-Based Testing in Resource-Constrained Environments." In 2025 22nd International Learning and Technology Conference (L&T), vol. 22, pp. 36-41. IEEE, 2025.
- Phaphuangwittayakul, Aniwat, Napat Harnpornchai, Fangli Ying, and Jinming Zhang. "RailTrack-DaViT: A Vision Transformer-Based Approach for Automated Railway Track Defect Detection." Journal of Imaging 10, no. 8 (2024): 192.
Aniwat Phaphuangwittayakul, Fangli Ying*, Yi Guo, Guohui, Surachai Santisookrat, "Adaptive Adversarial Prototyping Network for Few-Shot Prototypical Translation". Journal of Visual Communication and Image Representation 2023 (CCF C类期刊，SCI在线发表，if=2.887)[pdf]
Aniwat Phaphuangwittayakul, Fangli Ying*, Yi Guo, Liting Zhou, Nopasit Chakpitak, "Few-shot image generation based on contrastive meta-learning generative adversarial network". , 1-14, The Visual Computer 2022 (CCF C类期刊，SCI在线发表，if=2.601)[pdf]
F. Ying* , A.Y., Phaphuangwittayakul, A.#, Yi G. "Meta-FAVAE: Toward Fast and Diverse Few-shot Image Generation via Meta-Learning and Feedback Augmented Adversarial VAE". ICLR 2022 DGM workshop [ICLR 2022 Conference workshop]
Lebedeva, I.,F. Ying*, Guo, Y.* "Personalized facial beauty assessment: a meta-learning approach", The Visual Computer , (2022). (CCF C类期刊，SCI在线发表，if=2.601) [pdf]
Lebedeva, I., Guo, Y.* & F. Ying, "MEBeauty: a multi-ethnic facial beauty dataset in-the-wild", Neural Computing and Application, Jun. 2021, 195, pp. 116595. (CCF B类期刊，SCI在线发表，if=5.606) [pdf][code]
A.Y., Phaphuangwittayakul, A.#, Yi G., F. Ying#, "Fast Adaptive Meta-Learning for Few-shot Image Generation", IEEE Transaction on Multimedia, Nov. 2021, 124, pp. 308-314. (CCF B类期刊，SCI在线发表，if=5.452) [pdf][code]
A. Phaphuangwittayakul, Y. Guo, F. Ying*, "An optimal deep learning framework for multi-type hemorrhagic lesions detection and quantification in head CT images for traumatic brain injury", Applied Intelligence, Dec. 2021, 209, pp. 106478. (CCF B类期刊，SCI收录，if=5.086) [pdf]
A. Phaphuangwittayakul#, Y. Guo, F. Ying#, W. Xu and Z. Zheng, "Self-Attention Recurrent Summarization Network with Reinforcement Learning for Video Summarization Task," in 2021 IEEE International Conference on Multimedia and Expo (ICME), Shenzhen, China, 2021 pp. 1-6. （CCFB类会议） [pdf][code]
Dawod, A.Y.#, Phaphuangwittayakul, A.#, F. Ying#, Angkurawaranon S., Chakpitak N., Adaptive Slices Brain Hemorrhage Segmentation based on SLIC Algorithm, In 2020 17th Engineering Letter, 2020（EI数据库）
Lebedeva, I; Guo, Y; F. Ying Transfer Learning Adaptive Facial Attractiveness Assessment Journal of Physics: Conference Series; Bristol Vol. 1922, Iss. 1, (May 2021).（EI数据库）
Note: * indicates the corresponding author. # indicates equal contribution.
Executable Opensource Tutorial Books for Lectures in CS
- I created this book for the third-year undergraduate course of ‘Introduction to Engineering’
- : I created this book for the first-year postgraduate course of ‘Deep Learning’
Academic service
PC & TC
Reviewer
TMM, TIP, ICLR2024, ACMMM 2025, IJCNN2025, JCML, NACC, IEEE SPL, ECAI2024, ICDIP, WACV2021， BMVC2021, 计算机工程, Scientific reports
Patents
[1]Ying, Fangli, Luo, Ziyue, and Li, Yade. "A Multimodal Video Summary Generation Method Based on Temporal Alignment and Semantic Enhancement." CN119202313B. 2025-03-28. (Licensed)
[2]Ying, Fangli, Fan, Ru, and Li, Yade. "An Adaptive Stock Investment Decision Generation Method Suitable for High - Volatility Markets." CN119579319A. 2025-03-07.
[3]Ying, Fangli, Luo, Ziyue, and Li, Yade. "A Multimodal Video Summary Generation Method Based on Temporal Alignment and Semantic Enhancement." CN119202313A. 2024-12-27.
[4]Zhang, Jing, Gao, Yongwei, Yuan, Yubo, Chen, Zhihua, Ying, Fangli, and Jiang, Lei. "A Method for Removing Over - segmented Edges of Digital Images Based on Color Similarity." CN104680539B. 2018-11-09. (Licensed)
[5]Yuan, Yubo, Dai, Guanghui, Liu, Yun, Chen, Zhihua, Zhang, Jing, and Ying, Fangli. "An Automatic Content Detection and Repair Method for Scribbled Digital Images." CN104361565B. 2018-10-23. (Licensed)
[6]Huang, Zhigang, and Ying, Fangli. "A Virtual Reality Glasses." CN206788473U. 2017-12-22. (Licensed)
[7]Liu, Kebin, and Ying, Fangli. "A Visible Toothbrush." CN206151615U. 2017-05-10. (Licensed)
[8]Chen, Zhihua, Jin, Zhengtae, Liang, Jianning, Yuan, Yubo, Zhang, Jing, and Ying, Fangli. "A Real - Time Gesture Recognition Method Based on Finger Segmentation." CN104063059B. 2017-01-04. (Licensed)
[9]Yuan, Yubo, Dai, Guanghui, Chen, Zhihua, Zhang, Jing,Ying, Fangli, and Liu, Yun. "An Image and Video Information Hiding Method Based on Block Scrambling and Non - linear Transformation." CN104822067A. 2015-08-05. (Licensed)
[10]Yuan, Yubo, Liu, Yun, Chen, Zhihua, Zhang, Jing, Ying, Fangli, and Dai, Guanghui. "A Chinese Character Image Scaling Method Based on Bilinear Operators." CN104700357A. 2015-06-10. (Licensed)
[11]Zhang, Jing, Gao, Yongwei, Yuan, Yubo, Chen, Zhihua, Ying, Fangli, and Jiang, Lei. "A Method for Removing Over - segmented Edges of Digital Images Based on Color Similarity." CN104680539A. 2015-06-03. (Licensed)
[12]Chen, Zhihua, Jiang, Lei, Yuan, Yubo, Zhang, Jing, Ying, Fangli, and Zhou, Liqihan. "A Digital Image Restoration Method Based on Scale Optimization." CN104680493A. 2015-06-03. (Licensed)
[13]Zhang, Jing, Li, Da, Yuan, Yubo, Chen, Zhihua, Ying, Fangli, and Feng, Shengwei. "An Image Semantic Annotation Method Based on Multi - layer Segmentation." CN104636761A. 2015-05-20. (Licensed)
[14]Yuan, Yubo, Liu, Yun, Dai, Guanghui, Chen, Zhihua, Zhang, Jing, and Ying, Fangli. "An Image Foreground Extraction Method Based on Gaussian Variation Model." CN104484665A. 2015-04-01. (Licensed)
[15]Yuan, Yubo, Dai, Guanghui, Liu, Yun, Chen, Zhihua, Zhang, Jing, and Ying, Fangli. "An Automatic Content Detection and Repair Method for Scribbled Digital Images." CN104361565A. 2015-02-18. (Licensed)
[16]Chen, Zhihua, Jin, Zhengtae, Liang, Jianning, Yuan, Yubo, Zhang, Jing, and Ying, Fangli. "A Real - Time Gesture Recognition Method Based on Finger Segmentation." CN104063059A. 2014-09-24. (Licensed)
Projects
Participated in the research of "Efficient Exploration and Intelligent Manufacturing of Natural Products through Multi - source and Composite Approaches" as Co-PI (National Key R & D Program, Ministry of Science and Technology of the People's Republic of China, Project No. 2020YFA0907800), October 2021 - November 2025, Funding: 14.13 million yuan
Participated in the research of "Research on Key Technologies of Speech Emotion Recognition Integrating Brain - mechanism Learning" (General Program, National Natural Science Foundation of China, 62276098), January 2023 - December 2026
Participated in the research of "Research on Video Restoration Technology Based on Lightweight Meta - operator Deep Neural Network" (General Program, National Natural Science Foundation of China, 62272164), January 2023 - December 2026
Participated in the research of "Development of Intelligent Light - controlled Microbial Reactor" as Co-PI (National Major Scientific Research Instrument Development Project, National Natural Science Foundation of China, Project No. 32327801), January 1, 2024 - December 31, 2028, Funding: 7.9935 million yuan, Ongoing
Industrial Bioprocessing, March 2021 - Present, Funded by the State Key Laboratory of Bioreactor Engineering at ECUST
Constructed a multi-variety time series data monitoring system for large-scale cell culturing.
Developed a generative neural network to optimize sequence fitness and diversity.
Designed an enzyme with optimal Michaelis constants based on structural features.
Reinforcement Learning for Portfolio Management, June 2021 - Present, in collaboration with the Department of Finance at ECUST
Proposed a deep machine learning solution to the portfolio management problem.
Co-supervised a PhD student whose research topic was Portfolio Management.
Adapted deep reinforcement learning methods to the dynamic cryptocurrency market.
Developed a deep learning feature extractor for rotation strategies.
Smart City Project for Shanghai, June 2017 - February 2019, Funded by the National Engineering Laboratory for Big Data Distribution and Exchange Technologies
Created the industrial park GIS map for the government's smart city platform.
Formulated an e-commerce strategy for the Nanjing Road Shopping Area.
Analyzed downtown house prices by integrating the spatial weighted regression model.
Smart Chat Generation for English Online Education, September 2020 - June 2021, Funded by World Foreign Language Education in JuneYao Co. Ltd
Offered NLG and TTS solutions for smart chat generation.
Integrated a VAE for customized prosody feature generation in TTS.
Studied the generation of multiple accurate and diverse chat responses for the same query in short-text conversation tasks.
Data-driven knowledge system for Smart Customer Service Assistant, May 2016 - February 2017, Funded by KKH Global
Developed e-commerce SaaS projects for top brand companies.
Developed a saliency detection-based system to enhance user visual experience in e-commerce tasks.
Previous Experience during Ph.D
Worked as a GIS Expert at the World Bank and published GIS data visualization from the Climate Change Center of Central Asia.
Worked as a PhD candidate in Computer Science and collaborated on a project with the National Center of Geocomputation (NCG) in Ireland.
Teaching
Ph.D. Supervision
A.Y., Phaphuangwittayakul Ph.D. Thesis: Deep Generative Models for Few-Shot Image Generation and its Applications
Graduated, Now Teaching in ChiangMai UniversityIrina Lebedeva Ph.D. Thesis: Computational Facial Attractiveness: Assessment and Synthesis in Generic and Personalized Scenarios
Graduated,Now Working in Zhijiang Laboratory
Teaching in Computer Science
An opensource excutable textbook for Deep Learning course : Introduction to Engineering
- 2014-2016, Information Security for Postgrads, Advances in Computer Science for Undergrads, Artificial Intelligence for Adult Education
- 2016-2018, Information Security for Postgrads, Image Processing and Computer Vision for Undergrads, processing programming language for Art student, Multi-Media Technology and Introduction to logic design for Adult Education, Practice of Computer Science for 1st Year students, Artificial Intelligence for cross model students
- 2018-present, Image Processing and Computer Vision for Undergrads, Practice of Computer Science for 1st Year students, Introduction to logic design for international student in 2021, Summer School Seminar for the students of The University of Dundee, Summer School Seminar for the students of Chiang Mai University, MBA course of digital innovation (1 class)
UnderGraduate Projects
- 2023, 3D Human Body Reconstrcution with Multi-View Images For Metaverse, Students Research Program
- 2021, A Study and Application of Video Editing Techniques Based on 3D Deep Convolutional Networks, Students Research Program (1 Software Patent)
- 2020, Financial Data Visualization System Based on Quantitative Strategy Models, Students Research Program (1 Software Patent)
- 2019, Development of an AI-Based Automatic Assessment System for Teaching Chinese as a Foreign Language, Students Research Program (1 Software Patent)
Education
Ph.D., Computer Science, Maynooth University, 09.2013
Thesis: Management of spatial data for visualization on mobile devices
B.Sc., Software Engineering, Zhejiang University, 06.2009
Main Courses: C Programming, Embedded Systems, Computer Network and Communication, Software Engineering.
Competitions and awards
Best Runner, in ACM-GIS , 2011
ESRI Scholar, in AGILE conference 2010
Work experience updated by 2021
Lectuer in Computer Science in ECUST, 12.2013-Present
Developing the newest AI algorithms and work with Financial Department and Biology Department for industrial problems
Developing the Computational Human Visual Systems
supervising 3 international Ph.D ; supervising many postgraduates and undergraduates
Teaching Image processing, computer vision, information security, AI, introduction to computer science ;
Visiting Profosser, in ICDI in CMU, 02.2019-Present
Instructed two undergraduate and 5 Ph.D students in Digital Innovation
Tracked, studied, reproduced, and improved up-to-date Digital Innovation methods
Published papers on Digital Innovation
Research Scientist, in KKH, 07.2017-06.2018
Working as Chief Scientist in KKH global company and leading the team to develop the E-commence SaaS projects for the Top brand companies
Built a Visual System for User Experience Design for Top E-commence brands
Built a User Comments Management SaaS Systems for Top E-commence brands
Research Assistant, World Bank, 04.2013-09.2013
Data Visualization and Climate Changing Data Publication for Center Asia.