Jerome Leudet
Papers
2
Total Citations
5
H-Index
2
About
Jerome Leudet is a researcher at the forefront of computer vision and deep learning, with a focused expertise in camera calibration and geometric image understanding. His most significant contribution is the development of "Deep-BrownConrady," a pioneering deep learning framework that predicts camera calibration and distortion parameters from a single image. This work directly addresses a long-standing challenge in the field—traditionally requiring physical targets or multiple views—by demonstrating that a neural network, trained on a strategic mix of real and synthetic data, can achieve accurate, single-shot parameter estimation. This breakthrough has immediate implications for robotics, augmented reality, and autonomous systems, where rapid and robust calibration is critical. While his work is recent, with his 2025 paper already accumulating over 5 citations, the innovative methodology and practical utility signal a high-impact trajectory. Leudet’s research effectively bridges the gap between synthetic data generation and real-world deployment, offering a scalable solution that reduces the need for cumbersome calibration procedures. His work stands as a key reference for researchers exploring learning-based approaches to geometric computer vision problems.
Research Focus
Key Achievements
Top Papers
- 1
- 2