Saksham Bhutani

Bangalore, Karnataka

India

I am a graduate student in the Department of Electrical and Computer Engineering at Carnegie Mellon University. I hold a background in electronics engineering from Vellore Institute of Technology (VIT) in Vellore, India.

My research journey began under the guidance of Dr. Shunmuga Perumal at the Automotive Research Center at VIT, where I focused on developing perception models for autonomous vehicles operating in challenging environments with poor road infrastructure. I then interned at ETH Zurich, working on extracting PPG data from facial videos using computer vision, under the mentorship of Dr. Moe Elgendi and Dr. Carlo Menon. Following this, I worked as a Pre-Doctoral Fellow at the Indian Institute of Science (IISc), collaborating with Dr. Vaibhav Katewa on deep learning-based Camera-LiDAR fusion techniques for mapping and navigation. Additionally, I served as a Visiting Researcher at Khalifa University, where I worked again with Dr. Moe Elgendi on developing intelligent biomedical devices.

I have also had the privilege of leading the Avionics and Software departments at Assailing Falcons, a team that achieved 3rd place Worldwide and 1st place in Asia in the SAE Aero Design Competition in 2021-2022. I also founded and chaired the VIT Amateur Radio Club (VARC).

In my free time, I love building and flying drones, capturing breathtaking footage that you can find here. I’m also a licenced amateur radio operator (Callsign: VU3DTU) and an avid traveler. Per aspera ad astra 💫

news

Jan 01, 2025	I am starting graduate studies at Carnegie Mellon University.
Nov 01, 2024	I am working as a Visiting Researcher at Khalifa University for two months.
Nov 01, 2023	I have received Pre-doctoral Fellowship from the Kotak-IISc AI Center (KIAC).

selected publications

Journal
LaneScanNET: A deep-learning approach for simultaneous detection of obstacle-lane states for autonomous driving systems

P Shunmuga Perumal^†, Yong Wang, M Sujasree , and 4 more authors

Expert Systems with Applications, 2023

Abs Bib HTML

This paper proposes a deep-learning-based novel decision-making network named LaneScanNET to assist the ADS in lane-changing or lane-keeping decision-making. The proposed system assists ADS in detecting obstacles, localization of the Ego Vehicle (EGV) on roads, and estimating lane status in its Field of View (FOV). The proposed LaneScanNET uses a parallel pipeline with an Obstacle Detection Network (ODN) and a Lane Detection Network (LDN) to simultaneously process the incoming image frames for detecting obstacles and segmenting lane lines, respectively. Further, the Obstacle-Lane Fusion Network (OLFN) fuses these results to predict the status of the obstacle lane in the FOV of autonomous vehicles. Vellore Institute of Technology’s (VIT’s) real-time driving dataset on indigenous roads has been collected to train the proposed LaneScanNET with 2464 obstacle-lane images obtained by driving over 60 km. The dataset generated for obstacle detection and its corresponding lane detection has also been made publicly available to promote research work for these developing countries. The proposed system outperforms all existing networks on Indian roads with an accuracy of 75.28% in obstacle detection and 91.36% in lane detection. Additionally, the proposed system can analyze the lane status with an accuracy of 92.54%. The proposed network performs exceptionally well in unforeseen circumstances like shadows, fog, dust, and occlusions making LaneScanNET a robust network that can be an add-on for ADS to make lane-keeping or lane-changing decisions. The LaneScanNET can be integrated into real-time vehicles to assist drivers or ADS.
@article{perumal2023lanescannet, title = {LaneScanNET: A deep-learning approach for simultaneous detection of obstacle-lane states for autonomous driving systems}, author = {Perumal, P Shunmuga and Wang, Yong and Sujasree, M and Tulshain, Shobhit and Bhutani, Saksham and Suriyah, M Kiruthik and Raju, V Umesh Kumar}, journal = {Expert Systems with Applications}, volume = {233}, pages = {120970}, year = {2023}, publisher = {Elsevier}, doi = {https://doi.org/10.1016/j.eswa.2023.120970}, dimensions = {true}, }