Bharat Lal Bhatnagar

University of Tübingen, TH Bingen University of Applied Sciences

Papers

Total Citations

184

H-Index

About

Bharat Lal Bhatnagar is a computer vision and 3D scene understanding researcher whose work centers on modeling the complex dynamics between humans and objects in real-world environments. His most significant contribution, the BEHAVE dataset and framework (2022), has rapidly become a foundational resource in the field, amassing over 146 citations and establishing a new benchmark for tracking human-object interactions in natural settings. BEHAVE addresses the critical challenge of generalizing across a vast range of objects and interaction scenarios, with direct implications for gaming, virtual and mixed reality, human behavior analysis, and human-robot collaboration. Building on this foundation, Bhatnagar extended his research with "Visibility Aware Human-Object Interaction Tracking from Single RGB Camera" (2023), which tackles the practical limitation of inconsistent 3D relative translation in single-camera reconstruction — a key barrier to deploying such systems in robotics and graphics pipelines. Together, these works reflect a coherent research vision: enabling machines to robustly perceive and reconstruct the 3D interplay between people and their surroundings using accessible, minimal sensor setups. His contributions are already shaping how the research community approaches embodied AI and interactive scene reconstruction.