Prospective annotation work in image and video tagging (bounding boxes, segmentation, keypoints) with QA
You annotate images using semantic segmentation and related visual tagging to produce consistent labeled outputs. You focus on capturing visual patterns and maintaining recurring character features to support reliable video/image generation workflows. You validate label quality by checking for inconsistencies and errors prior to dataset finalization. • Semantic segmentation for detailed region-level labeling • Bounding box-style tagging and key point annotation (where applicable) • Visual pattern identification for consistent labels • QA checks for labeling errors and distortions