Pierre-Yves Richard
Papers
2
Total Citations
14
H-Index
2
About
Pierre-Yves Richard is a computer vision researcher whose work centers on real-time camera relocalization—a critical technology for augmented reality and robot navigation. His major contributions lie at the intersection of machine learning and geometric approaches, where he has developed hybrid methods that achieve both speed and accuracy in challenging environments. Richard is best known for introducing "Accurate Sparse Feature Regression Forest Learning," a novel framework that combines sparse feature regression with random forest learning to enable robust, real-time camera pose estimation. He also pioneered "xyzNet," a scene coordinate prediction network that leverages deep learning to directly regress 3D coordinates from 2D image inputs, significantly improving localization performance in dynamic or texture-poor scenes. Each of these foundational papers has garnered 7 citations, reflecting their early influence on the field. Richard’s work is particularly notable for addressing the persistent trade-off between computational efficiency and localization precision, making his methods highly relevant for practical AR and robotics applications. His research continues to shape the development of lightweight, learning-based solutions for visual localization.
Research Focus
Key Achievements
Top Papers
- 1
- 2