Author ORCID: orcid.org/0000-0002-1352-9910
A system for estimating 3D human joint locations from monocular images.
@inproceedings{nibali2019margipose,
title={3D Human Pose Estimation with 2D Marginal Heatmaps},
author={Nibali, Aiden and He, Zhen and Morgan, Stuart and Prendergast, Luke},
booktitle={2019 IEEE Winter Conference on Applications of Computer Vision (WACV)},
pages={1477-1485},
year={2019},
doi={10.1109/WACV.2019.00162}
}
A comparison of approaches to predicting keypoint locations from images using convolutional neural networks.
@article{nibali2018numerical,
title={Numerical Coordinate Regression with Convolutional Neural Networks},
author={Nibali, Aiden and He, Zhen and Morgan, Stuart and Prendergast, Luke},
journal={arXiv preprint arXiv:1801.07372},
year={2018}
}
A video analysis system for athletic diving with large deep learning components. The model is able to extract action clips, track athletes, and classify the type of dive performed.
@inproceedings{nibali2017extraction,
author = {Nibali, Aiden and He, Zhen and Morgan, Stuart and Greenwood, Daniel},
title = {Extraction and Classification of Diving Clips From Continuous Video Footage},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
pages={38--48},
month = {July},
year = {2017}
}
Application of a deep learning classifier to subjective malignancy prediction of lung cancer nodules. The model is able to approximate the predictive power of a human radiologist ensemble.
@article{nibali2017pulmonary,
title={Pulmonary nodule classification with deep residual networks},
author={Nibali, Aiden and He, Zhen and Wollersheim, Dennis},
journal={International Journal of Computer Assisted Radiology and Surgery},
pages={1--10},
year={2017},
publisher={Springer International Publishing}
}
A compression algorithm designed for GPS trajectory data. The algorithm is able operate in lossy and lossless modes.
@article{nibali2015trajic,
title={Trajic: An effective compression system for trajectory data},
author={Nibali, Aiden and He, Zhen},
journal={IEEE Transactions on Knowledge and Data Engineering},
volume={27},
number={11},
pages={3138--3151},
year={2015},
publisher={IEEE}
}
RatTrace is the intelligent rat trap monitoring system which I developed for my final year undergraduate engineering project. It works by measuring rat bait levels and wirelessly reporting the results so that pest control experts don’t need to physically check each trap on location.
1000WAT is a Thai food restaurant based in Melbourne’s CBD. I created a new website for 1000WAT with a focus on responsive design and simple content management. Ruby on Rails was used on the server-side, with jQuery and Twitter Bootstrap providing a good client-side user experience.
This robot is able to balance itself on two wheels. The project involved laying out a PCB using Altium Designer, soldering the manufactured circuit board, and writing the embedded C code required to make it balance.
The seed-dropping robot was created as an entrant for the 2014 National Instruments Autonomous Robotics Commpetition. This robot was able to navigate an obstacle course unassisted, using feedback from a LiDAR and wheel encoders to determine an appropriate course.
This project was developed by a team of students from La Trobe University, and was a finalist in the 2014 NI ARC competition.
Home Guardian is a home monitoring system designed for people who live alone and are at an increased risk of falling victim to an injury or illness in and around the home that would otherwise affect their ability to seek help on their own.
This project was developed by a team of students from La Trobe University, and won 1st place in Telstra’s inaugral M2M University Challenge in 2013.
MediCom is a real-time communication system designed for use in the medical sector. It makes use of WebRTC for video conferencing capabilities and runs on both mobile devices and PCs.
This project was developed by a team of students from La Trobe University according to a brief from Ericsson Australia.