AI & ML Engineer - Image Annotation and Bounding Box Correction
ImageBounding Box
Audited and relabeled over 500 images to correct bounding box errors, enhancing the reliability of a computer vision model for smart attendance. Focused on reducing false positives by ensuring precise annotation and consistent labeling standards. Collaborated in stabilizing the curriculum codebase for high-volume student usage, directly impacting model performance.
• Conducted thorough review and correction of bounding box placements on image datasets.
• Improved annotation guidelines and maintained data quality standards.
• Utilized OpenCV and Mediapipe tools in the annotation workflow.
• Ensured alignment between relabeled data and model training objectives.
Audited and relabeled over 500 images to correct bounding box errors, enhancing the reliability of a computer vision model for smart attendance. Focused on reducing false positives by ensuring precise annotation and consistent labeling standards. Collaborated in stabilizing the curriculum codebase for high-volume student usage, directly impacting model performance. • Conducted thorough review and correction of bounding box placements on image datasets. • Improved annotation guidelines and maintained data quality standards. • Utilized OpenCV and Mediapipe tools in the annotation workflow. • Ensured alignment between relabeled data and model training objectives.
2020 - 2020