Shivam Duggal

Papers

Total Citations

H-Index

About

Shivam Duggal is a researcher advancing the frontiers of 3D computer vision, with a core focus on neural implicit modeling for object reconstruction. His work tackles one of the field’s most persistent challenges: generating high-fidelity 3D shapes from sparse, partial, and real-world data—a critical capability for autonomous driving, robotics, and augmented reality. In his highly cited 2022 paper, “Mending Neural Implicit Modeling for 3D Vehicle Reconstruction in the Wild” (24 citations), Duggal identifies and remedies the fundamental limitations of existing neural implicit methods when applied to sparse, single-view observations, proposing novel techniques to achieve robust, detailed reconstructions. His earlier foundational work, “Secrets of 3D Implicit Object Shape Reconstruction in the Wild” (8 citations), systematically dissects why state-of-the-art methods fail on real-world, cluttered scenes and offers principled solutions to bridge the gap between synthetic benchmarks and practical deployment. By explicitly addressing the “in-the-wild” domain shift, Duggal’s contributions have provided a crucial roadmap for making 3D reconstruction reliable outside controlled lab settings, directly impacting applications in autonomous perception and interactive graphics. His research stands as a vital reference for any student or engineer seeking to build 3D systems that work robustly in the messy, unpredictable real world.