Sivaji Retta

I am an AI Engineer specializing in Machine Learning and Computer Vision, passionate about crafting cutting-edge solutions to complex real-world problems. Holding a B.Tech in Electronics and Communication Engineering from Rajiv Gandhi University of Knowledge Technologies (RGUKT-IIIT), Nuzvid, India (2021), my technical expertise spans Machine Learning, Deep Learning, Computer Vision, Image Processing, and Time Series Analysis. A collaborative professional by nature, I thrive on knowledge-sharing and teamwork to accelerate AI innovation. My career journey includes leading the development of advanced ML and vision-based systems at SENTINL, contributing to interdisciplinary AI initiatives as a Visiting Researcher at the Singapore University of Technology and Design (SUTD), designing and deploying production-grade vision models at AnimalEYEq, building scalable ML pipelines for enterprise clients at Quantiphi Analytics Solutions, and most recently, delivering data-driven insights and Computer Vision solutions for the manufacturing industry as a Data Scientist at Capgemini Engineering in the Middle East, Dubai.

Download CV | Email | Google Scholar | LinkedIn | GitHub

Research

CattleEyeView: A Multi-task Top-down View Cattle Dataset for Smarter Precision Livestock Farming

KE Ong, Sivaji Retta, R Srinivasan, S Tan, J Liu
2023 IEEE International Conference on Visual Communications and Image Processing (VCIP), 2023
IEEE / Google Scholar
We propose CattleEyeView, a comprehensive dataset for precision livestock farming, focusing on multi-task learning in a top-down view setting, improving cattle monitoring and management.

Comprehensive Analysis of Deep Learning Approaches for PM2. 5 Forecasting

Sivaji Retta, P Yarramsetti, S Kethavath
Proceedings of International Conference on Computational Intelligence and Data Engineering (ICCIDE), 2020
Springer / Google Scholar
This paper provides a detailed analysis of various deep learning methodologies for forecasting PM2.5 levels, addressing critical environmental challenges through advanced AI techniques.

Towards Imperceptible Adversarial Image Generation: Minimizing Perceptual Difference

Sivaji Retta, R Srinivasan
25th IEEE International Symposium on Multimedia, 2023
IEEE / Google Scholar
This research introduces novel techniques for generating imperceptible adversarial images, focusing on minimizing perceptual differences to enhance image security in digital media.

MTYOLO: A Multi-Task Model to Concurrently Obtain the Vital Characteristics of Individuals or Animals

KE Ong, Sivaji Retta, R Srinivasan, S Tan, J Liu
2024 IEEE International Conference on Multimedia and Expo Workshops (ICMEW), 2024
IEEE / Google Scholar
MTYOLO presents a multi-task model aimed at concurrently determining vital characteristics in both individuals and animals, leveraging advanced deep learning techniques.

CattleDeSegNet: A Joint Approach to Cattle Denoising and Interpretable Segmentation

Sivaji Retta, R Srinivasan, S Tan
16th SPIE International Conference on Machine Vision, 2023
SPIE Digital Library Google Scholar
CattleDeSegNet introduces a joint model that combines cattle denoising and interpretable segmentation, addressing key challenges in livestock monitoring through AI.

DebrisVision: Bridging the Synthetic-to-Real Gap for Enhanced Underwater Debris Analysis

Sivaji Retta, SME Machara, II Ganapathi
ICCV 2025 Workshop on Computer Vision for Autonomous Underwater Intervention and Marine Vision (CVAUI-AAMVEM)
LinkedIn Post
DebrisVision introduces a large-scale dataset featuring 9,430 real-world and 15,570 synthetic underwater images with multi-modal annotations. This work bridges the synthetic-to-real domain gap to advance underwater debris detection, robotic navigation, and ecological monitoring.

Professional Experience

Data Scientist

Capgemini Engineering

Mar 2025 - Present

Dubai, UAE

AI Researcher

SENTINL

Apr 2024 - March 2024

Singapore

Visiting Researcher

SUTD

Jan 2024 - Jun 2024

Singapore

AI Researcher

animalEYEQ

Oct 2022 - Apr 2024

Singapore

ML Engineer

Quantiphi

May 2021 - Oct 2022

Mumbai, India

Research Intern

IIT Hyderabad

Feb 2020 - Mar 2020

Hyderabad, India

Research Intern

NIT Rourkela

May 2019 - Jul 2019

Rourkela, India

Demo Videos

Animal Weight Estimation

Animal Counting

Yolo-NAS Object Detection

Yolo-NAS Pose estimation

SAM 2

YoloV8 key Point detection

ByteTrack - Player Tracking

BotSort - Player Tracking

OCSORT- Player Tracking

InnoTrack - Retail Surveillance & Customer Insights

AlignFit - Perfecting your From with Innobayt AI

OCSORT- Player Tracking

Hackathons & Scholarship

SUS INNOHACK Hackathon

Open Innovation Hackathon

Fiction2Science Hackathon

Open Innovation Hackathon, KL University, India

2017, Winner

Awarded for Building Smart Villages in collaboration with Andhra Pradesh Innovation Society and UC Berkeley.

Pravega Innovation Summit, IISc Bangalore, India

2019, Finalist

Designed a live location tracking wristband, ranking among the top 8 finalists at the National Level Pravega Innovation Summit.

SUS INNOHACK Hackathon, RGUKT Nuzvid, India

2019, Winner

Received a winning prize of Rs. 25,000 at the hackathon organized by Rajiv Gandhi University of Knowledge Technologies and INIAC.

Fiction2Science Hackathon, Continental, Bangalore, India

2019, Finalist

Recognized as one of the top 8 innovative teams for developing a Deep Learning model for pre-crash detection.

Sankalp Semiconductor Hackathon, Kolkata, India

2020, Finalist

Designed a workflow for Automated and Decentralized Pollution Monitoring and Forecasting using IoT and ML, winning the competition.

Fully-funded Integrated Program, India

Top 5%

Awarded a fully-funded six-year Integrated Program, encompassing Pre-University Course and Bachelor of Technology, as one of the top 5% of rural secondary school graduates in India.

Merit Scholarship

Secondary Education

Awarded Merit Scholarship during Secondary education.

Travel Gallery

I love exploring new places and cultures. So far, I've traveled to India, Singapore, Malaysia, and Thailand, with many more countries on my list for the future.