I’m an Australian PhD researcher in artificial intelligence and computer vision at the Australian National University in Canberra, Australia. My supervisors are Dylan Campbell, Stephen Gould, and Hongdong Li. My research focuses on 3D reconstruction from limited data (creating 3D models from 2D images), basically helping computers see and understand the world in three dimensions, even when they don’t have all the pieces of the puzzle.
My LinkedIn is here, and my GitHub is here.
I’m also involved in AI safety research, particularly evaluating large language models using algorithmic information theory. I’m always interested in meeting new people and hearing about potential collaborations. If you’d like to get in touch with me, please email me at sam [dot] bahrami [at] anu [dot] edu [dot] au.
📚 Work
First PhD publication PlückeRF: A Line-based 3D Representation for Few-view Reconstruction at CVPR
AI Safety Research UK AI Safety Institute bounty recipient for developing a novel evaluation method for large language models using algorithmic information theory
Co-author The Edge of Disaster: A Battle Between Autonomous Racing and Safety
💼 Experience
PhD Researcher at the Australian National University (2023-present)
Machine Learning and Research Engineer (2019-2023)
Australian Institute for Machine Learning and Australian Centre for Robotic Vision working with Ian Reid, Tat-Jun Chin and Anton Van Den Hengel
🌍 Places & Engagements
CVPR 2025 presenting PlückeRF: A Line-based 3D Representation for Few-view Reconstruction (photo)
AMSI Summer School 2025 enrolled in Elliptic PDEs from an Elementary Viewpoint by Enrico Valdinoci, and Optimal transportation and Monge-Ampère equations by Jiakun Liu (photo)
Research @ Sydney Google Event 2025
Australian AI Safety Forum 2024 (photo)
ANU Hackathon Healthcare and AI themed, second place (photo)
Made a public art vending machine for the city of Adelaide (photo)
Learn2Race Challenge Virtual racing competition winners, paper published at ICML
Developer on the NASA Robotics Challenge won a $100k USD prize
Govhack Hackathon winners