Machine Learning and Software Development Projects
I'm an Computer Vision Engineer with a passion for deep learning. Some of the my projects I've worked on are shown below.
At KEF Robotics, I've led the development of a pipeline for creating "Digital Twins" of objects through the use of Gaussian Splatting.
At KEF Robotics, I've fine-tuned monocular depth estimation networks and applied them to autonomous flight. Recently a new SOTA network from TikTok was released called Depth-Anything. I can't share my work projects but I did make a huggingface space to try the Depth-Anything model on your own video files!
Monocular Depth Estimation PyTorchLess than 24 hour hackathon for tech in healthcare. SkinsAI stands for Skin Key Imaging Nursing System, and it's a free-access diagnosis tool for classifying moles as benign or malignant. SkinsAI's binary classification model uses a small convolutional neural network. The purpose is to provide a free, easy-to-use, anonymous and fast tool to give patients a first diagnosis, as well as a website where useful and up-to-date information can be gathered. We were awarded 2nd place overall out of 29 teams ($2,000 cash prize).
PyTorch Django Computer VisionCompetition for predicting the aqueous solubility class (high, medium, low) of small molecules. I applied a graph neural network called SolTranNet to solve the problem.
PyTorch