Sahil Jatoi

Sahil Jatoi

Student Researcher
MINES Lab (Mobile INtelligent Embedded System)
Department of AI and Robotics
Sejong University, South Korea

Masters → PhD Candidate in AI & Robotics

Specializing in Vision AI, Deep Learning & Computer Vision

Google Scholar GitHub Email LinkedIn

News

[2025] Started Masters → PhD in AI & Robotics at Sejong University

[2025] Joined MINES Lab as Student Researcher focusing on Vision AI

[2025] Paper submitted to Scientific Reports Journal (Nature)

[2024] Paper "FaultSeg: A Dataset for Train Wheel Defect Detection" published in Scientific Data Journal (Nature)

[2024] Submitted multiple papers to GCWOT2024 Conference (IEEE)

[2024] Joined as AI Engineer at NCRA-CMS Lab, MUET

[2023] Graduated with BE in Computer Systems Engineering (CGPA: 3.15)

About Me

I am currently a Student Researcher at MINES Lab (Mobile INtelligent Embedded System), Department of AI and Robotics, Sejong University, Seoul, South Korea.

Previously, I worked as an AI Engineer at the National Center of Robotics and Automation (NCRA-CMS Lab) at MUET, where I deployed AI models into mobile and web applications and embedded AI models into robotics machines. I completed my Bachelor's in Computer Systems Engineering from Mehran University of Engineering & Technology.

My research interests focus on:

Selected Publications
FaultSeg paper
FaultSeg: A Dataset for Train Wheel Defect Detection

Sahil Jatoi, et al.
Scientific Data Journal (Nature), 2024

[paper] [dataset]

Smog Prediction paper
Harnessing Machine Learning for Accurate Smog Level Prediction: A Study of Air Quality in India

Sahil Jatoi, et al.
VAWKUM Transactions on Computer Sciences, 2025

[paper]

Road Maintenance paper
Towards Smarter Road Maintenance: YOLOv7-Seg for Real-Time Detection of Surface Defects

Sahil Jatoi, et al.
Pattern Recognition. ICPR 2024

[paper]

Predictive Maintenance paper
Predictive Maintenance in Urban Railway Systems Using Machine Learning Models

Sahil Jatoi, et al.
GCWOT2024 Conference (IEEE)

[paper]

Key Projects
Train Wheel Detection
Real-time Train Wheel Defect Detection System

Developed a comprehensive system using object detection and instance segmentation for real-time identification of train wheel defects. Deployed on mobile applications for field use.

Tech: Python, Deep Learning, YOLOv8, Computer Vision

Education App
Multilingual Image Recognition for Early Education

Built an interactive learning application supporting English, Urdu, and digits with CNN-based image recognition, integrated with Google's voice library for pronunciation.

Tech: Python, CNN, Streamlit, TensorFlow

Robotics RL
Reinforcement Learning for Industrial Robotics

Integrated reinforcement learning in industrial pick-and-place robotics and autonomous robotic arm grasping.

Tech: Python, Reinforcement Learning, ROS