John Power

Key Skills

Software

Labelbox

Top Subject Matter

No subject matter listed

Top Data Types

Image

Video

Top Label Types

Bounding Box

Point Key Point

Segmentation

Classification

Tracking

Emotion Recognition

Freelancer Overview

I am an experienced AI Training Specialist with a strong background in data labeling and annotation for image, video, and audio datasets. Over the past three years, I have supported computer vision and speech recognition projects by preparing high-quality training data using tools like Labelbox, CVAT, and YOLO. My expertise includes object detection, multi-object tracking, frame-by-frame video annotation, and audio transcription. I am skilled in managing annotation workflows, conducting quality assurance checks, and collaborating with machine learning engineers to refine guidelines and improve model performance. My attention to detail and commitment to data quality have enabled me to deliver tens of thousands of accurately labeled samples, driving improvements in model accuracy across a variety of AI projects.

Labeling Experience

AI-Powered Multi-Object Detection & Video Tracking Annotation Project

LabelboxVideoBounding BoxPoint Key Point

Currently working on a large-scale video annotation project focused on training and improving computer vision models for object detection and tracking applications. Responsible for frame-by-frame video labeling, including bounding boxes, polygon segmentation, and persistent object ID assignment to ensure accurate multi-object tracking across sequences. Using Labelbox and CVAT to manage annotation workflows, maintain dataset organization, and ensure consistency across thousands of video frames. Supporting model validation processes using YOLO to evaluate detection accuracy and refine annotation quality. Key responsibilities include: Multi-object tracking in dynamic and real-world environments Handling occlusions, fast motion, and overlapping objects Maintaining high annotation precision (98%+ accuracy) Conducting quality assurance and peer review checks Collaborating with ML engineers to improve labeling guidelines This ongoing project contributes to AI systems used in surveilla

Currently working on a large-scale video annotation project focused on training and improving computer vision models for object detection and tracking applications. Responsible for frame-by-frame video labeling, including bounding boxes, polygon segmentation, and persistent object ID assignment to ensure accurate multi-object tracking across sequences. Using Labelbox and CVAT to manage annotation workflows, maintain dataset organization, and ensure consistency across thousands of video frames. Supporting model validation processes using YOLO to evaluate detection accuracy and refine annotation quality. Key responsibilities include: Multi-object tracking in dynamic and real-world environments Handling occlusions, fast motion, and overlapping objects Maintaining high annotation precision (98%+ accuracy) Conducting quality assurance and peer review checks Collaborating with ML engineers to improve labeling guidelines This ongoing project contributes to AI systems used in surveilla

2024

AI-Powered Multi-Object Detection and Video Tracking Annotation Project

LabelboxImageBounding BoxPoint Key Point

Led a large-scale data annotation project supporting computer vision model development for object detection and tracking systems. Annotated over 50,000+ images and 2,000+ video clips using bounding boxes, segmentation masks, and frame-by-frame tracking techniques. Utilized Labelbox to manage dataset workflows and ensure annotation consistency across teams. Applied object detection validation processes using YOLO to improve training accuracy. Performed detailed multi-object tracking in video datasets and labeled audio files for speech recognition model training. Maintained a 98%+ annotation accuracy rate through strict quality control and double-review systems. Collaborated closely with machine learning engineers to refine labeling guidelines, reduce bias in datasets, and improve overall AI model performance.

Led a large-scale data annotation project supporting computer vision model development for object detection and tracking systems. Annotated over 50,000+ images and 2,000+ video clips using bounding boxes, segmentation masks, and frame-by-frame tracking techniques. Utilized Labelbox to manage dataset workflows and ensure annotation consistency across teams. Applied object detection validation processes using YOLO to improve training accuracy. Performed detailed multi-object tracking in video datasets and labeled audio files for speech recognition model training. Maintained a 98%+ annotation accuracy rate through strict quality control and double-review systems. Collaborated closely with machine learning engineers to refine labeling guidelines, reduce bias in datasets, and improve overall AI model performance.

2021 - 2024

Education

Work History