Madhur Gupta

Vellore Institute of Technology University

Papers

Total Citations

H-Index

About

Madhur Gupta is a researcher in robotics and computer vision, with a focus on deep learning for autonomous manipulation. His most notable contribution lies in developing learning-based approaches for robotic grasping, particularly through his work on "Object Grasping using Convolutional Neural Networks" (2019). In this study, Gupta demonstrated how a robotic arm can autonomously grasp objects or clear clutter by leveraging a pre-trained AlexNet architecture on ImageNet. By continuously feeding visual data into a deep neural network, his system enables real-time identification and manipulation of objects, bridging the gap between perception and action in robotics. Though his citation count is modest, his work represents an early and practical application of transfer learning to robotic grasping—a foundational step for more advanced systems today. Gupta’s research is especially valuable for students and engineers interested in integrating convolutional neural networks with physical robotic control, offering a clear pathway from simulation to real-world deployment. His contributions highlight the growing synergy between deep learning and robotics, paving the way for more intelligent and adaptive automation.