Xinyue(Katrina) LI
Sydney,NSW
Summary
Sophisticated Machine Learning Engineer with background in independent research of end-to-end optimized image compression.
Demonstrated success in designing novel ML-based approaches to compression problems, and developing prototypes to assess viability of approach.
Expert with documented history in aligning with other departments and patent attorney for successful patent filing.
Overview
1
1
year of professional experience
Work History
Post-doctoral Research Associate
The University Of New South Wales, Sydney
2023.03 - Current
- Implemented and evaluated different machine learning algorithms and neural network architectures for improved learned wavelet-like image compression.
- Designed and executed different training strategies to model the discontinuous quantization and entropy coding processes, for the back-propagation of end-to-end optimized image encoding frameworks.
- Contributed to and participated heavily in creating provisional and formal international patents under the Patent Cooperation Treaty (PCT).
- Collaborated with other departments and patent attorney to facilitate successful patent filing.
- Leveraged interpersonal and communication skills to mentor graduate and undergraduate students.
- Contributed dominantly to manuscript writing process for publication of research papers.
- Published research results in top ranking machine-learning journals (TPAMI), and presented at seminars and conference meetings.
- Maintained accurate records of research findings and provided statistical analysis of data results.
Education
Ph.D. - Machine Learning For Image Compression
The University Of New South Wales, Sydney
UNSW SYDNEY, NSW, 205602.2023
Master of Science - Signal Processing (Image Compression)
The University Of New South Wales, Sydney
UNSW SYDNEY, NSW, 205601.2019
Bachelor of Science - Electronics And Information Engineering
Guangzhou University
Guangdong, China06.2016
Skills
- Strong proficiency in Python
- Deep understanding and experience in Tensorflow/Keras and Pytorch
- Deep knowledge in common machine-learning algorithms, e.g. CNN, RNN and GAN
- Practical experience in patent filing under the PCT
- Outstanding verbal and written communication skills
- Extremely organized and self-motivated
Patent
1. TAUBMAN, David Scott; NAMAN, Aous Thabit; LI, Xinyue, “Method, apparatus and computer readable medium for encoding an image,” Australian Provisional Patent Application No. 2022902529, Filed on 2 September 2022.
2. TAUBMAN, David Scott; NAMAN, Aous Thabit; LI, Xinyue, “Method, apparatus and computer readable medium for encoding an image,” International Patent Application under the PCT, No. P0050162PCT, Filed on 29 August 2023.
Publications
1. Xinyue Li, Aous Naman and David Taubman, "Neural Network Assisted Lifting Steps For Improved Fully Scalable and Lossy Image Compression In JPEG2000", IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), under review.
2. Xinyue Li, Aous Naman and David Taubman, "Studies On Learning-based Lifting Structures For Fully Scalable and Accessible Lossy Image Compression", IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), under review.
3. Xinyue Li, Aous Naman and David Taubman, "Improved Transform Structures For Learned Wavelet-like Fully Scalable Image Compression", The IEEE International Workshop on MultiMedia Signal Processing (MMSP 2023).
4. Li, X., Naman, A. and Taubman, D., 2022, October. "A Neural Network Lifting Based Secondary Transform for Improved Fully Scalable Image Compression in Jpeg 2000". In 2022 IEEE International Conference on Image Processing (ICIP)(pp. 1606-1610). IEEE.
5. Li, X., Naman, A. and Taubman, D., 2021, September. "Machine-Learning Based Secondary Transform for Improved Image Compression in JPEG2000". In 2021 IEEE International Conference on Image Processing (ICIP) (pp. 3752-3756). IEEE.
6. Li, X., Naman, A. and Taubman, D., 2020, October. "Adaptive secondary transform for improved image coding efficiency in jpeg2000". In 2020 IEEE International Conference on Image Processing (ICIP) (pp. 1216-1220). IEEE.
Timeline
Post-doctoral Research Associate
The University Of New South Wales, Sydney
2023.03 - Current
Ph.D. - Machine Learning For Image Compression
The University Of New South Wales, Sydney
Master of Science - Signal Processing (Image Compression)
The University Of New South Wales, Sydney
Bachelor of Science - Electronics And Information Engineering
Guangzhou University
Similar Profiles
Kesewa AgyemangKesewa Agyemang
Research Assistant at University Of New South Wales, SydneyResearch Assistant at University Of New South Wales, Sydney
Syed Abbas Hasan ZaidiSyed Abbas Hasan Zaidi
Egineering Project Assistant at University of New South WalesEgineering Project Assistant at University of New South Wales
Yuqian LinYuqian Lin
Marketing at University of New South WalesMarketing at University of New South Wales