Neural Radiance Fields (NeRF) for 3D Product Generation
NeRF-based 3D product visualisation pipeline built at Deep Design Systems. Combines neural radiance fields with diffusion model conditioning to generate photorealistic 3D …
•
1 min read
LLM Fine-tuning & RAG Pipelines
Fine-tuning large language models using LoRA and QLoRA for domain-specific tasks, with retrieval-augmented generation (RAG) pipelines for grounded outputs. Evaluated and refined …
•
1 min read
Aerodynamic Shape Optimisation with Deep Learning
Surrogate deep learning model replacing computationally expensive CFD simulations for car aerodynamics shape optimisation. Trained on CFD-generated flow field data to predict drag …
•
1 min read
CT Organ Segmentation in Cancer Patients
Deep learning pipeline for 3D organ segmentation in oncology CT scans, used in clinical research at Radboud University Medical Center. Trained on large annotated datasets of cancer …
•
1 min read
Transfer learning from a sparsely annotated dataset of 3D medical images
Transfer learning from a sparsely annotated dataset.
Kidney abnormality segmentation in thorax-abdomen CT scans
Transfer learning from a sparsely annotated dataset.
The Liver Tumor Segmentation Benchmark (LiTS)
Automatic detection and segmentation of battery cells of Electric Vehicles.
Patrick Bilic
EV Battery Cell Segmentation via CT
Automated 3D segmentation of lithium-ion battery cells from industrial CT scans, deployed in production at Nuctech. Pipeline built with TensorRT and ONNX for optimised inference, …
•
1 min read
Automatic Segmentation of Battery Cells of Electric Vehicles
Automatic detection and segmentation of battery cells of Electric Vehicles.
Winning Brazil's National COVID-19 AI Challenge
How I built an automated CT scan segmentation system that won 1st place in Brazil's government-sponsored national AI competition.