Data Labeler / Video Annotator — Atlas Capture (Remote)
Performed dense video labeling and frame-by-frame action tagging to support computer vision and AI training workflows. Mapped hand-object interactions and reviewed output to ensure consistency with detailed project guidelines and edge-case requirements. Maintained high accuracy benchmarks expected for machine learning datasets. • Dense video labeling across frames • Frame-by-frame action tagging • Hand-object interaction mapping • Quality checks against changing guidelines