Victor N.

Key Skills

Software

Data Annotation Tech

Top Subject Matter

Computer vision gesture recognition and action labeling (hand tracking and gestures).

Gesture recognition and action detection video dataset annotation and quality correction.

Computer vision video frame annotation for object detection and human activity recognition.

Top Data Types

Video

Image

Text

Top Task Types

Tracking

Action Recognition

Object Detection

Freelancer Overview

Senior Video Annotation Specialist — DataVision AI Solutions (Remote). Core strengths include Internal, Proprietary Tooling, and N. Education includes Bachelor of Science, University of Nigeria (2020). AI-training focus includes data types such as Video and labeling workflows including Tracking, Action Recognition, and Object Detection.

Labeling Experience

Senior Video Annotation Specialist — DataVision AI Solutions (Remote)

VideoTracking

Edited and corrected AI-generated video annotations at scale with emphasis on frame-level timestamp synchronization, bilateral hand tracking, and multi-class action labeling. Reviewed in-house annotation model outputs to refine labels, reduce error rates, and improve average accuracy to 97.4% using systematic QA protocols. Authored and updated annotation guidelines, including standardized timestamp rounding conventions that improved team-wide consistency by 25%. • Maintained 6–8 annotated video hours per day to meet pipeline targets for training dataset delivery. • Collaborated with annotation leads and ML engineers to align on evolving labeling standards and resolve edge cases in complex multi-hand gestures. • Performed quality benchmarking and edge case detection during daily annotation QA. • Produced documentation and feedback loops to continuously improve labeling processes.

2023 - Present

Video Annotation Analyst — MotionTag Data Services (Remote)

VideoAction Recognition

Performed high-volume correction of video annotations for gesture recognition and action detection datasets, focusing on fine-grained hand pose and motion labeling. Processed an average of 4–5 hours of annotated video per day while logging work hours for client SLA throughput requirements. Identified recurring left/right hand classification errors and proposed label correction macros to reduce annotation time per clip by 20%. • Participated in weekly calibration sessions to align annotation quality across a 12-person remote team. • Delivered project batches with client-side quality acceptance above 96%. • Conducted dataset output error identification and recurring-issue remediation. • Supported guideline clarification for ambiguous scenarios via written feedback.

2021 - 2022

Junior Data Annotator — CrowdLabel Technologies (Remote)

VideoObject Detection

Annotated video frames for object detection and human activity recognition tasks, building foundational expertise in temporal labeling and action segmentation. Ranked among the top 10% of annotators by accuracy and speed metrics on the platform leaderboard. Assisted senior annotators by reviewing flagged edge cases and contributing to internal best-practice documentation for timestamp annotation. • Supported edge case review and quality improvement through collaboration with senior annotators. • Contributed to knowledge base articles covering timestamp annotation best practices. • Performed frame-level labeling to support downstream computer vision training datasets. • Improved throughput and labeling accuracy based on platform performance metrics.

2020 - 2021

Education

Work History