Cloud Credits
Received $ 10K cloud credits from Amazon Web Services (AWS) for RootNous
Received initial $10K cloud credits for RootNous, for research and development in AI and related fields. Thanks to Amazon for the support.
AI researcher · Full-stack engineer · Founder
Building intelligent systems at the seam of research and product.
I work on computer vision, remote sensing, robotics, bioinformatics and more - contributing open research, shipping AI systems that serve real users and businesses, and founding companies that turn research insights into useful products.
03
Peer-reviewed & preprint publications
02
Companies founded — RootNous, OriginSci
06+
Research domains explored
Scroll to exploreworld model· drag to explore
01 / About
Founder — RootNous & OriginSci
Researcher who publishes.
Engineer who codes.
Founder who ships.
Chirantan Ghosh is the founder of RootNous and OriginSci, where he drives both foundational and applied research in computer vision, machine learning, deep learning, remote sensing, and robotics. His work investigates how machines can perceive, interpret, and act within the physical world - spanning from fine-grained analysis of earth-observation imagery to the control of autonomous systems.
Before establishing RootNous, Chirantan focused on optimizing machine learning algorithms and published peer-reviewed studies on earth data. His interests lie in perceptual pattern recognition combined with structured reasoning, and in bridging rigorous research with deployable solutions across environmental, medical, and autonomous-systems domains. He is committed to open publication and to transforming research breakthroughs into commercial products when discoveries are ready to move beyond the lab.
Building
Deep-tech products
Publishing
Foundational & applied research papers in AI and related fields
Filing
Patents for AI and similar solutions
Research interests
Computer Vision·Remote Sensing·Robotics·Bioinformatics·Machine Learning·Deep Learning·Earth Observation·Autonomous Systems·Perceptual AI·Applied Research
02 / News
Latest updates.
News, milestones, and releases — straight from the desk.
Paper Review
Reviewed an Applied Deep Learning article for the SN Computer Science journal (Springer)
Reviewed it on Jan 2026, and was accepted for publication in August 2026. Thanks to the Editor and Chief Editor for this opportunity.
Visit Orcid Peer review sectionCloud Credits
Received $ 5K cloud credits from Microsoft for RootNous
Received initial $5K cloud credits for RootNous, for research and development in AI and related fields. Thanks to Microsoft for the support.
Company
Founded OriginSci — an open science initiative in AI and related fields
Building open infrastructure and open publication culture for AI research, from zero to one.
Visit OriginSciVirtual Accelerator
RootNous got accepted into the Nvidia Inception program
The Nvidia Inception program is a virtual accelerator program that supports startups in AI and data science. Thanks to Nvidia for the support.
Preprint
Released the first version of the preprint “Review of Key Image Denoising Algorithms”
A review and comparative analysis of Gaussian, linear, and non-linear isotropic smoothing for noise reduction.
Read preprintApp
Okexpert App went live on Apple store (No longer available on Apple store)
It is a Marketplace of Experts, Influencers and Organizations, where users can search and book an expert from any field around the world for an exclusive video call session.
App
Okexpert App went live on Android play store
It is a Marketplace of Experts, Influencers and Organizations, where users can search and book an expert from any field around the world for an exclusive video call session.
Visit Okexpert AppCompany
Founded RootNous — research & development in AI and related fields
Applied and foundational research in computer vision, machine learning, deep learning, remote sensing, and robotics.
Visit RootNousArticle
Paper published in Geoscientific Model Development (GMD)
A methodological framework for improving the performance of data-driven models — Sobol sensitivity analysis with Bayesian optimization, demonstrated on daily runoff prediction in the Maumee domain, USA.
View publicationArticle
Paper published in Geophysical Research Letters (GRL)
Generalization of runoff risk prediction at field scales to a continental-scale region using cluster analysis and hybrid modeling.
View publication03 / Publications
Selected research.
Peer-reviewed and preprint work in machine learning, earth observation, and environmental modeling.
[01]Preprint
Preprint · 2025
Review of Key Image Denoising Algorithms
Chirantan Ghosh
Images are one of the key sources of visual information and communication. It plays a crucial role in defense, AI, and forensic science, among others. However, it is prone to corruption from various types of noise from varying sources, mainly during acquisition and transmission. This paper reviews the existing techniques and analyzes the performance of three main techniques — Gaussian, linear, and non-linear isotropic smoothing — finding that both linear and non-linear smoothing can be effective solutions.
[02]Article
Geoscientific Model Development · 2023
A methodological framework for improving the performance of data-driven models: a case study for daily runoff prediction in the Maumee domain, USA
Yao Hu*, Chirantan Ghosh*, Siamak Malakpour Estalaki
Because of the black-box nature of data-driven models, their performance cannot be guaranteed. We developed a generalizable framework combining hyperparameter selection based on Sobol global sensitivity analysis with hyperparameter tuning via Bayesian optimization — demonstrated through daily edge-of-field runoff predictions using XGBoost in the Maumee domain, USA. The framework contributes towards improving the performance of a variety of data-driven models.
[03]Article
Geophysical Research Letters · 2022
Generalization of Runoff Risk Prediction at Field Scales to a Continental-Scale Region Using Cluster Analysis and Hybrid Modeling
Chanse M. Ford, Yao Hu, Chirantan Ghosh, Lauren M. Fry, Siamak Malakpour Estalaki, Lacey Mason, Lindsay Fitzpatrick, Amir Mazrooei, Dustin C. Goering
We develop a regionalization approach based on principal component analysis and K-means clustering to identify clusters with similar runoff potential over the Great Lakes region. For each cluster, hybrid models combine NOAA's National Water Model with XGBoost and field-scale measurements — enabling prediction of daily runoff risk level at the field scale over the entire region.
04 / Projects
Open source & products.
Companies built from zero to one — turning research insights into useful products.
Product · Founder · 2026 — present
OriginSci
Open Science Initiative in AI and related fields.
0 → 1·AI·ML·Deep Learning·Computer Vision·Robotics·Remote Sensing·Applied Research·Research & Development
Visit edu.africa
Product · Founder · 2023 — present
RootNous
R&D in AI and related fields.
0 → 1·AI·ML·Deep Learning·Computer Vision·Robotics·Remote Sensing·Applied Research·Research & Development
Visit rootnous.com
05 / Contact
Building something intelligent?
Open to research collaborations, advisory roles, and the occasional founding conversation. If you're working on something hard at the seam between research and product, I'd love to hear about it.