Abhirup Das

Papers

Total Citations

H-Index

About

Abhirup Das is a researcher whose work lies at the intersection of computer vision, robotics, and human-object interaction. His primary focus is on developing algorithms that enable machines to understand and manipulate objects in the physical world, with a particular emphasis on grasp-pose prediction—a critical challenge for robotic manipulation and augmented reality. In his most-cited paper, "Grasp-Pose Prediction for Hand-Held Objects" (2019), Das introduced a novel approach to inferring how a hand might grasp an object based solely on its visual appearance, moving beyond traditional object-centric models to consider the dynamic relationship between hand and object. This contribution has garnered 5 citations, laying a foundation for more intuitive human-robot collaboration. Das’s work is notable for its practical implications: by predicting plausible grasps, his research directly supports advancements in assistive robotics, prosthetics, and interactive AI systems. His approach bridges the gap between perception and action, offering a pathway for robots to handle everyday objects with human-like dexterity. For students and researchers, Das’s research exemplifies how targeted, application-driven work in computer vision can solve real-world manipulation problems, making him a promising voice in the field of embodied AI.