CV

Summary Download PDF

I am a Computer Science PhD Researcher at QUT Centre for Robotics, Australia. I am expected to finish by the end of 2021.  Currently I am interning as an Applied Scientist at Amazon.

My research is in the field of computer vision with the focus on deep learning with limited annotated data. I am passionite applying machine learning for real-life applications. My research has been used to build and deploy deep neural networks for applications in wildlife conservation.

Work experience

• Jul 2021 – Present: Applied Scientist Intern

  Amazon, Australia

  Achievements:

  • applied machine learning, computer vision, data mining and statistical techniques to improve customer experience,
  • developed, prototyped and scaled machine learning models.

  Technologies: PyTorch, AWS

• Sep 2020 – Jun 2021: ML Engineer, contractor

  WildMe/Wildbook, Portland, Oregon, USA

  Achievements:

  • implemented and delivered AI models for wildlife detection and identification,
  • integrated the solution as a plugin to existing infrastructure.

  Technologies: PyTorch, OpenCV, open source GitHub

  Source code: orientation network, re-identification network

• Jul 2018 – Nov 2019: Tutor (Advanced AI course)

  Queensland University of Technology, Brisbane, Australia

  Achievements:

  • prepared course materials and assignments,
  • presented lectures and tutorials on building AI models,
  • marked assignments and exams.

• Oct 2016 – Sep 2017: Research Assistant, part time

  Central Queensland University, Brisbane, Australia

  Achievements:

  • implemented the multi-objective optimisation algorithm and the genetic algorithm in Java for the problem of generating a nutritionally balanced diet,
  • delivered a prototype of a smart dietary system that generates daily meals schedules based on dietary requirements.

• Jun 2009 - May 2014: SAP ERP Consultant, full time

  Several consulting companies based in Moscow, Russia

  Achievements:

  • analysed and designed business processes in logistics for clients in energy and gas mining industries,
  • customised SAP system to tailor for business requirements,
  • prepared project documentation,
  • liaised with stakeholders and key end-users.

Education

• Ph.D in Computer Science, Queensland University of Technology, Australia, 2021 (expected)
• M.S. in Information Systems, Central Queensland University, Australia, 2016
• B.S. in Mathematics, Lomonosov Moscow State University, Russia, 2009

Skills

• Research skills:
  • designing and building AI models for computer vision tasks;
  • contribution to research submissions in top AI conferences;
  • writing research papers and technical reports.
• Technical skills:
  • proficiency in Python including Pytorch, Keras, OpenCV, SciPy, Numpy, Pandas;
  • deployment of AI models and web applications (AWS, MS Azure, Docker).
  • practical experience in building, deploying and testing ML models in a product development context using software engineering best practices;
• Soft skills:
  • communication with non-technical stakeholders;
  • presentation skills.

Publications

Learning geometric equivalence between patterns using embedding neural networks

Olga Moskvyak & Frederic Maire (2017). "Learning geometric equivalence between patterns using embedding neural networks." In Proc. International Conference on Digital Image Computing: Techniques and Applications (DICTA).

Robust re-identification of manta rays from natural markings by learning pose invariant embeddings

Olga Moskvyak, Frederic Maire, Asia O Armstrong, Feras Dayoub & Mahsa Baktashmotlagh (2021). "Robust re-identification of manta rays from natural markings by learning pose invariant embeddings." In Proc. International Conference on Digital Image Computing: Techniques and Applications (DICTA).

Learning Landmark Guided Embeddings for Animal Re-identification

Olga Moskvyak, Frederic Maire, Feras Dayoub and Mahsa Baktashmotlagh. (2020). "Learning Landmark Guided Embeddings for Animal Re-identification." In Proc. Winter Conference on Applications of Computer Vision Workshops

Keypoint-aligned embeddings for image retrieval and re-identification

Olga Moskvyak, Frederic Maire, Feras Dayoub, Mahsa Baktashmotlagh. (2021). "Keypoint-aligned embeddings for image retrieval and re-identification." In Proc. Winter Conference on Applications of Computer Vision (WACV). pp. 676-685.

Semi-supervised keypoint localization

Olga Moskvyak, Frederic Maire, Feras Dayoub & Mahsa Baktashmotlagh (2021). "Semi-supervised keypoint localization." In Proc. International Conference on Learning Representations (ICLR).

Going Deeper into Semi-supervised Person Re-identification

Olga Moskvyak, Frederic Maire, Feras Dayoub & Mahsa Baktashmotlagh (2021). "Going Deeper into Semi-supervised Person Re-identification." Arxiv preprint, In Review.

Learning from limited annotated data for re-identification problem