Self-Supervised Learning for 3D Multimodal Perception – Funded PhD opening
Do you want to work for two of the world’s leading research organisations, live in a city with extremely high standard of living, enjoy pristine beaches and sun all year round, and still carry out ground breaking Machine Learning, Robotics research to solve real-world challenges? If so, this position is for you.
The Robotics and Autonomous Systems Group at CSIRO’s Data61 have been developing the state of the art multi-modal-based 3D SLAM systems that are able to be used for driverless navigation, mapping, scene understanding and manipulation of loads and objects. This can be done in industrial, urban or natural environments.
A Fully Funded PhD position is available as a part of a research collaboration between the Robotics and Autonomous Systems group at the Commonwealth Scientific and Industrial Organization (CSIRO) and the Queensland University of Technology (QUT), in Brisbane, Australia. You will receive a scholarship of $28,000 per year for 3.5 years. Top-up of $10,000 per year will be awarded to outstanding students.
Topic: Self-Supervised Learning for 3D Multimodal Perception
Potential impact of deep learning is limited due to the lack of large, annotated, and high-quality datasets in domains of interest. Annotating such datasets is laborious, costly and time-consuming. This project proposes to develop self-supervised learning systems to extract and use the relevant context given by strong prior spatio-temporal models (e.g. dense 3D reconstructions) as supervisory signals in training. This new concept will investigate model structures that encodes spatio-temporal data, and show rapid adaptation of models to new domains (few-shot learning) using trained embeddings layers (self-supervised, or prior data).
The PhD student will leverage from CSIRO expertise in 3D multimodal perception with QUT deep knowledge of deep learning in vision to develop algorithms that could extract and use the relevant context given by CSIRO in-situ 3D systems as supervisory signals in training.
- Must have a Bachelor’s degree with the first Class Honours or a Master’s degree with Research in a relevant area in the past 5 years (e.g., Computer Science, Electrical Engineering, Mechatronics, Physics or other related fields)
- Strong competencies in one or more of the followings areas: Robotics, Computer Vision, Machine Learning, and Deep Learning.
- Demonstrated strong programming skills in C++ or Python in Linux.
- Demonstrated Research Experience e.g. a good publication record.
- Demonstrated Experience in Robot Operating System, Tensorflow and/or Pytorch.
How to apply
Prospective students should send the following documents in a SINGLE PDF file to Dr. Peyman Moghadam (email@example.com) with the subject [PhD SSL], including:
- a current c.v.
- details of grades or an academic transcript
- one page cover letter explaining your research background and interests,
Dr. Peyman Moghadam (firstname.lastname@example.org)