大叔 子.

Key Skills

Software

CVAT

Label Studio

LabelImg

Top Subject Matter

Autonomous Driving

Medical and General Text Data Labeling

General Image Labeling and OCR

Top Data Types

Image

Text

Audio

3D Sensor

Top Task Types

Bounding Box

Entity (NER) Classification

Classification

Emotion Recognition

Transcription

Freelancer Overview

Data Labeling Specialist, Images and 3D Data. Core strengths include CVAT, Label Studio, and LabelImg. Education includes Bachelor of Science, 厦门理工学院 (2021). AI-training focus includes data types such as Image, Text, and Audio and labeling workflows including Bounding Box, Entity (NER) Classification, and Classification.

Labeling Experience

Data Labeling Specialist, Images and 3D Data

CVATImageBounding Box

Responsible for 2D and 3D labeling of autonomous driving images and point cloud data, including lane lines, traffic signs, pedestrians, and vehicles. Used project guidelines to annotate continuous frames and refined predictions from pre-annotation models. Maintained a personal data inspection accuracy rate above 98% and regularly participated in quality review meetings.• Conducted detailed object annotation for autonomous driving scenes.• Maintained high standards in data quality and labeling consistency.• Utilized pre-labeling model outputs for time-efficient manual correction.• Collaborated with team lead to refine unclear labeling rules and feedback difficult cases.

2024 - Present

Data Labeling Specialist, Text and Medical Imaging

Label StudioTextEntity (NER) Classification

Performed entity recognition and relationship extraction annotation for text, and handled some image classification and semantic segmentation tasks. Consistently delivered over 1,500 annotated text items daily, achieving a pass rate of 98.5% for medical image lesion segmentation. Assisted new team members with tool usage and rule explanation.• Labeled medical and general domain text entities and relationships.• Completed image classification and semantic segmentation on various tasks.• Checked and clarified uncertain samples for audit readiness.• Supported and trained five new team members on labeling standards.

2022 - 2023

Data Labeling Specialist, Images/OCR

LabelImgImageClassification

Worked on basic image labeling tasks, including image classification, 2D bounding box annotation, and OCR transcription. Regularly processed more than 1,500 images per day with an annotation accuracy rate of approximately 98%. Used tools such as LabelImg, LabelMe, and CVAT to complete tasks and conduct batch processing and format conversion.• Performed high-volume image annotation across a variety of tasks.• Ensured annotation and processing accuracy in a quality-controlled workflow.• Applied OCR transcription and basic preprocessing to improve data value.• Delivered results after self-review and timely incorporation of feedback.

2021 - 2022

Fusion Annotation Project, Intersections

CVATImageBounding Box

Participated in image and point cloud fusion scene labeling for autonomous driving intersections. Focused on bounding box annotation of vehicles and pedestrians based on merged image and sensor data. Worked collaboratively with the team to support reliable training data production.• Performed fusion annotation with point cloud and image datasets.• Labeled vehicle and pedestrian objects in intersection scenes.• Ensured dataset consistency for autonomous driving model development.• Participated in team quality checks and process improvements.

2021 - 2021

Audio Transcription Project, Voice Assistant Data

Label StudioAudioTranscription

Annotated and verified wake word/transcription data for in-car voice assistants across diverse accents and environments. Maintained accuracy under delivery pressure, providing over 12,000 audio labels by project completion. Ensured that voice assistant training data met project and quality expectations.• Transcribed and annotated wake word utterances under varied noise conditions.• Maintained rigorous standards for accuracy and consistency.• Worked with project-specific guidelines for in-car speech datasets.• Delivered required data volumes within tight timeframes.

2021 - 2021

Show more

Education

Work History