Video Annotator
I worked on many video annotation projects aimed at training a computer vision model for motion detection and activity recognition, labeling over 1,500 video clips. My role involved frame-by-frame annotation with bounding boxes, ensuring consistency with object IDs, and addressing challenges like occlusions and fast motion to maintain 97–98% accuracy, bolstered by regular quality audits. This experience enhanced my precision and focus in managing dynamic visual data.