Basavaraja Shanthappa Vandrotti

Papers

Total Citations

H-Index

About

Basavaraja Shanthappa Vandrotti is a computer vision researcher specializing in depth estimation and 3D scene understanding. His primary contributions lie in developing deep learning architectures for depth completion—the task of generating dense, accurate depth maps from sparse sensor data. His most cited work, "DeepDNet: Deep Dense Network for Depth Completion Task" (2021, 12 citations), introduces a novel dense network that transforms sparse depth inputs into high-quality dense maps, directly supporting critical applications in 3D reconstruction, mixed reality, and robotics. Building on this foundation, his more recent "DeepSmooth: Efficient and Smooth Depth Completion" (2023, 6 citations) addresses the challenge of producing spatially and temporally consistent depth maps, essential for advanced functionalities like Spatial Mapping and video portrait effects in AR systems. Vandrotti's research systematically tackles the trade-off between computational efficiency and output quality, making his methods practical for real-time deployment. His work is particularly impactful for students and engineers working on autonomous navigation, augmented reality, and any domain requiring robust environmental perception from limited sensor input.