High-Accuracy Autonomous Vehicle 2D/3D Bounding Box Annotation
Machine Learning Data Annotator | Autonomous Vehicle Navigation Annotated 15,000+ high-resolution images and video frames to train ML models for autonomous vehicle perception and navigation. Key Responsibilities: 1. Object Identification: Created precise 2D bounding boxes and 3D cuboids for pedestrians, vehicles, cyclists, and traffic signs. 2. Attributes Tagging: Labeled objects with sub-types, occlusion levels, and movement direction. 3. Quality Assurance: Maintained 98.5% accuracy against strict client gold-standard guidelines. 4. Tool Proficiency: Used CVAT and LabelImg for efficient data processing and formatting.