Avinash Kumar Singh

Key Skills

Software

LabelImg

Top Subject Matter

No subject matter listed

Top Data Types

Image

Top Label Types

Bounding Box

Object Detection

Text Summarization

Freelancer Overview

I have hands-on experience in data labeling and annotation, supporting AI and machine learning initiatives by delivering high-quality, structured datasets for training and validation. My work includes labeling and validating both image and text datasets, with a strong focus on accuracy and reliability—achieving over 95% accuracy in annotation tasks. I am skilled in performing data validation and conducting rigorous quality checks to ensure datasets are ready for supervised ML models. My background in computer science and experience with tools like Git, Android Studio, and QA processes for ML datasets allow me to contribute effectively to projects in domains such as computer vision and natural language processing. I am committed to maintaining data integrity and supporting the development of robust AI systems through meticulous annotation and validation practices.

Labeling Experience

Automatic Number Plate Recognition (ANPR) – Vehicle License Plate Dataset Annotation

LabelimgImageBounding BoxObject Detection

Annotated and validated large-scale vehicle image datasets for Automatic Number Plate Recognition (ANPR) system development. 1). Performed bounding box annotation on 3,000+ vehicle images to localize license plates under varied lighting and environmental conditions (day/night/rain). 2). Ensured precise box alignment for improved YOLO-based object detection training. 3). Conducted quality assurance checks to maintain >95% annotation accuracy. 4). Filtered low-quality and blurred images to improve dataset reliability. 5). Supported OCR model training by validating extracted alphanumeric plate text. 6). Collaborated with ML engineers to refine labeling guidelines and improve detection precision.

Annotated and validated large-scale vehicle image datasets for Automatic Number Plate Recognition (ANPR) system development. 1). Performed bounding box annotation on 3,000+ vehicle images to localize license plates under varied lighting and environmental conditions (day/night/rain). 2). Ensured precise box alignment for improved YOLO-based object detection training. 3). Conducted quality assurance checks to maintain >95% annotation accuracy. 4). Filtered low-quality and blurred images to improve dataset reliability. 5). Supported OCR model training by validating extracted alphanumeric plate text. 6). Collaborated with ML engineers to refine labeling guidelines and improve detection precision.

2024 - 2024

Education

Work History