Sparsh Garg

I am currently working as a Machine Learning Engineer in the Perception team at Lucid Motors, where I focus on advancing autonomous driving technologies within the ADAS domain. My work centers around developing robust perception systems that enable safer and more intelligent vehicles.

Previously, I pursued my Master’s degree at Carnegie Mellon University, specializing in computer vision and robotics. At the Robotics Institute’s Kantor Lab, under the guidance of Dr. George A. Kantor, I worked on advancing automation technologies for agriculture and physical sciences, with an emphasis on perception and manipulation pipelines for real-world applications.

I also worked as a Research Intern at Bosch Center for Artificial Intelligence (BCAI), Sunnyvale, where I co-developed a state-of-the-art model for monocular metric depth estimation from arbitrary cameras.

Prior to my graduate studies, I earned my undergraduate degree from Punjab Engineering College, India. I then gained professional experience at ExxonMobil in Bangalore, and later at CynLr (Cybernetics Laboratory), where I played a key role in building a stereo vision system from the ground up, spanning both hardware and software development.

My overarching goal is to contribute to the realization of truly autonomous systems. My interests lie at the intersection of computer vision, deep learning, and robotics, with a focus on building reliable perception systems that can drive the next generation of intelligent vehicles.

    email: sparsh913@gmail.com
    Office: CIC Robotics Institute (near Lab 2)

Publications

Depth Any Camera (DAC): A Framework for Zero-Shot Metric Depth Estimation Across Diverse Camera Types
Yuliang Guo*, Sparsh Garg*, S. Mahdi H. Miangoleh, Xinyu Huang, Liu Ren
(Accepted to CVPR 2025)
pdf project page abstract bibtex

            @inproceedings{guo2025depthanycamera,
              title={Depth Any Camera: Zero-Shot Metric Depth Estimation from Any Camera},
              author={Yuliang Guo and Sparsh Garg and S. Mahdi H. Miangoleh and Xinyu Huang and Liu Ren},
              booktitle={arXiv},
              year={2025}
            }

SplatSim: Zero-Shot Sim2Real Transfer of RGB Manipulation Policies Using Gaussian Splatting
Mohammad Nomaan Qureshi, Sparsh Garg, Francisco Yandun, David Held, George Kantor, Abhisesh Silwal
(Accepted to ICRA 2025) (Spotlight Presentation at CoRL MRM-D Workshop)
pdf project page abstract bibtex

Sim2Real transfer, particularly for manipulation policies relying on RGB images, remains a critical challenge in robotics due to the significant domain shift between syn- thetic and real-world visual data. In this paper, we propose SplatSim, a novel framework that leverages Gaussian Splatting as the primary rendering primitive to reduce the Sim2Real gap for RGB-based manipulation policies. By replacing tradi- tional mesh representations with Gaussian Splats in simulators, SplatSim produces highly photorealistic synthetic data while maintaining the scalability and cost-efficiency of simulation. We demonstrate the effectiveness of our framework by training manipulation policies within SplatSim and deploying them in the real world in a zero-shot manner, achieving an average success rate of 86.25%, compared to 97.5% for policies trained on real-world data.

@misc{qureshi2024splatsimzeroshotsim2realtransfer,
      title={SplatSim: Zero-Shot Sim2Real Transfer of RGB Manipulation Policies Using Gaussian Splatting}, 
      author={Mohammad Nomaan Qureshi and Sparsh Garg and Francisco Yandun and David Held and George Kantor and Abhisesh Silwal},
      year={2024},
      eprint={2409.10161},
      archivePrefix={arXiv},
      primaryClass={cs.RO},
      url={https://arxiv.org/abs/2409.10161}, 
}