Jyoti Kini

Papers

Total Citations

H-Index

About

Jyoti Kini is a researcher at the forefront of 3D computer vision and autonomous systems, with a primary focus on self-supervised learning for perception in sparse, unlabeled environments. Her most cited work, "Self Supervised Learning for Multiple Object Tracking in 3D Point Clouds" (2022), tackles the critical challenge of tracking multiple objects in LiDAR data without costly human annotations—a bottleneck for mobile robots and autonomous driving. By designing a neural network that learns robust representations from raw point clouds alone, Kini’s approach significantly reduces the need for manual labeling while achieving competitive tracking performance. This contribution has already garnered 6 citations, reflecting its timely impact on the autonomous driving community. Her research bridges the gap between practical deployment and data efficiency, offering scalable solutions for real-world perception. Kini’s work is notable for addressing the dual hurdles of sparse 3D data and annotation scarcity, positioning her as an emerging leader in self-supervised learning for robotics. For students and researchers, her methods provide a blueprint for advancing perception systems where labeled data is limited.