Publications

Last updated on May 16, 2025

You can also find my articles on my Google Scholar profile.

An Efficient Federated Transfer Learning Approach for Multi-UAV Systems

Published in NCC, 2025

We propose a novel multi-UAV-based federated transfer learning system which drastically reduces the computational burden, centralizes it, reduces bandwidth requirements, and makes it more secure.

Recommended citation: Vedang Bhupesh Shenvi Nadkarni, Sandeep Joshi and L. Rajya Lakshmi, "An Efficient Federated Transfer Learning Approach for Multi-UAV Systems," 2025 National Conference on Communications (NCC), New Delhi, India, 2025, pp. 1-6, doi: 10.1109/NCC63735.2025.10983600.
Download Paper

GLiDR: Topologically Regularized Graph Generative Network for Sparse LiDAR Point Clouds

Published in CVPR, 2024

We propose GliDR, a graph generative network regularized by 0-dimensional Persistent Homolgy to densify globally consistent static LiDAR pointclouds.

Recommended citation: Prashant Kumar, Kshitij Madhav Bhat, Vedang Bhupesh Shenvi Nadkarni and Prem Kalra, "GLiDR: Topologically Regularized Graph Generative Network for Sparse LiDAR Point Clouds" in 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 2024, pp. 15152-15161, doi: 10.1109/CVPR52733.2024.01435
Download Paper | Download Slides

Differentiable SLAM Helps Deep Learning-based LiDAR Perception Tasks

Published in BMVC, 2023

We design a SLAM Loss to train LiDAR based models.

Recommended citation: Prashant Kumar, Dheeraj Vattikonda, Vedang Bhupesh Shenvi Nadkarni, Erqun Dong, and Sabyasachi Sahoo, “Differentiable SLAM helps deep learning-based lidar perception tasks” in 34th British Machine Vision Conference (BMVC), Aberdeen, UK, BMVA Press, 2023, p. 822
Download Paper | Download Slides