Nishad Gothoskar

Papers

Total Citations

H-Index

About

Nishad Gothoskar is a researcher at the forefront of integrating probabilistic reasoning with computer vision and robotics, specializing in inverse graphics and 3D scene understanding. His work introduces a paradigm shift by treating 3D perception as a probabilistic inference problem rather than a deterministic one. In his highly-cited 2023 paper, "3D Neural Embedding Likelihood: Probabilistic Inverse Graphics for Robust 6D Pose Estimation," Gothoskar pioneers a framework that combines neural embeddings with probabilistic modeling to achieve robust 6D object pose estimation from 2D images—a critical capability for autonomous manipulation and augmented reality. His earlier 2020 work, "Learning a generative model for robot control using visual feedback," demonstrates his ability to bridge perception and action by formulating a generative model that infers robot states from visual feature observations, enabling more adaptive and reliable control. With a growing citation impact, Gothoskar’s contributions are laying the mathematical and algorithmic groundwork for machines that can reason about 3D scenes with uncertainty, making him a rising voice in the quest for truly intelligent, visually-guided robotic systems.