Marie‐Julie Rakotosaona
Papers
2
Total Citations
29
H-Index
2
About
Marie-Julie Rakotosaona is a rising star in computer vision and 3D reconstruction, whose work pushes the boundaries of single-image 3D modeling. Her research centers on inverse rendering, neural radiance fields (NeRF), and generative adversarial networks (GANs), with a focus on extracting shape, pose, and appearance from minimal visual input. Her most-cited paper, "Shape, Pose, and Appearance from a Single Image via Bootstrapped Radiance Field Inversion" (2023, 26 citations), introduces a pioneering method that couples NeRF with GANs to reconstruct detailed 3D scenes from a single view—a significant leap over prior work that relied on synthetic datasets. This contribution addresses a critical bottleneck in 3D vision: enabling robust, topology-agnostic reconstruction from real-world images. Her earlier version of this work (2022, 3 citations) laid the groundwork for this approach. Rakotosaona's impact lies in advancing practical 3D reconstruction, with potential applications in augmented reality, robotics, and digital content creation. Her innovative bootstrapping technique demonstrates a clear path from synthetic to real-world generalization, marking her as a key contributor to the next generation of 3D vision technologies.
Research Focus
Key Achievements
Top Papers
- 1
- 2