High Precesion Multimodal Data Annotation for Computer Vision Models
Executed high-accuracy video data labeling and annotation for large-scale computer vision and AI training projects. Tasks included drawing bounding boxes, polygons, and keypoints for object detection and classification, as well as performing frame-by-frame tracking in video sequences. Ensured labeling consistency and accuracy through structured QA workflows and adherence to project guidelines. Collaborated with AI engineers to optimize dataset quality for autonomous vehicle and multimodal machine learning applications.