Vijay Badrinarayanan

Papers

Total Citations

H-Index

About

Vijay Badrinarayanan is a leading researcher in computer vision and deep learning, with a focus on 3D scene understanding and geometric perception from monocular imagery. His work bridges the gap between 2D image analysis and full 3D reconstruction, tackling fundamental challenges in indoor and outdoor environment modeling. He is best known for pioneering the Deep Cuboid Detector, a groundbreaking approach that moves beyond traditional 2D bounding boxes to directly localize 3D cuboids in cluttered scenes using end-to-end deep learning—a method that has garnered 18 citations and reshaped how researchers approach object detection in three dimensions. Badrinarayanan further advanced the field with DeepPerimeter, a sophisticated pipeline that infers complete indoor boundary maps from posed monocular sequences by integrating robust depth estimation and wall segmentation, earning 9 citations for its practical applications in robotics and augmented reality. His work consistently demonstrates an ability to solve complex geometric problems through innovative deep architectures, making him a respected voice in the computer vision community. Badrinarayanan’s contributions have significant implications for autonomous navigation, indoor mapping, and scene understanding, establishing him as a key innovator in 3D deep learning.