Naitri Rajyaguru

I'm a Computer Vision Research Engineer at Zupt LLC, a company at the forefront of underwater autonomous solutions.

In my career, I've worked at UMD's PRG lab with Prof. Aloimonos, contributing to their diverse research like auto calibration, drone navigation in complex environments,and coded aperture for navigation.

I have experience tackling a wide range of challenges, including tasks such as camera pose estimation and calibration, neural model compression, 3D scene reconstruction, image and depth inpainting, and robot localization.

Email / CV / Twitter / Github / LinkedIn


Computer Vision Research Engineer Zupt LLC July 2023 - Current	Research Intern Ford Summer 2022	M.Eng Robotics University of Maryland, College Park 2021-2023	Research Associate Swayyatt Robots Feb 2021 - July 2021	B.Tech ECE GTU 2015-2019

News

Excited to submit CodedVO to the RAL Journal.

CodedVO: Coded Visual Odometry
Naitri Rajyaguru*, Sachin Shah*, Chahat Deep Singh, Cornelia Fermüller, Christopher Metzler†, Yiannis Aloimonos†
*, † Equal Contribution
[Website]

CodedVO, a novel monocular visual odometry method that leverages optical constraints from coded apertures to resolve scale ambiguity of monocular depth estimation.

Selected Projects

I'm interested in developing minimal cognitive architectures and tackling the challenges of building deployable Artificial Intelligent (AI) frameworks for resource-constrained robots.

Structure From Motion (SFM)
Report / Code

3D reconstruction of a scene and camera pose estimation.

Lottery Ticket Hypothesis in Low Data Regime
Report / Code

Optimized model precision through Iterative Magnitude Pruning on 5% of weights with a 1000-sample dataset, creating a generalizable and high-performance model for diverse computer vision applications

Point Painiting : Point Cloud Object Segmentation
Code

Semantic Segmentation for LIDAR & Camera using SegFormer.

Super pixel generation using SLIC and Image Segmentation
Code

Segmentation network using SLIC Superpixels as input.

Panorama Stitching
Report / Code

Panoramic Image stitching using classical and deep learning method

Website Template Credit Link.