Eram Salva Nasreen

Key Skills

Software

Scale AI

SuperAnnotate

Label Studio

CVAT

Don't disclose

Top Subject Matter

No subject matter listed

Top Data Types

3D Sensor

Image

Text

Top Task Types

Tracking

Point Key Point

Entity Ner Classification

Segmentation

Bounding Box

Freelancer Overview

I hold a Master’s degree in Electronics and have a strong foundation in computer applications, including MS Word and basic computer operations. My academic background has given me the analytical and detail-oriented mindset required for data labeling and annotation roles. I am eager to apply my skills to support AI training data initiatives, ensuring accurate and high-quality data preparation for machine learning projects. I am quick to learn new technologies and applications, and I am committed to contributing to innovative projects in data annotation, whether in computer vision, natural language processing, or other domains. I am fluent in English, Hindi, Urdu, and Kannada, which enables me to work effectively in diverse teams and on multilingual datasets.

Labeling Experience

Body Keypoints (Human Pose Estimation) | QA Specialist

SuperannotateImagePoint Key Point

Directed the final QA review and annotation of 2D hand and body keypoints for driver and passenger analysis. Maintained rigorous standards for anatomical accuracy to support models in gesture recognition and in-cabin safety monitoring. Identified and corrected pose estimation errors to improve the reliability of human-computer interaction (HCI) models. • Ensured high fidelity in label placement for human pose estimation. • Supported HCI and in-cabin safety models through granular labeling. • Reduced error rates via systematic QA processes. • Enhanced gesture recognition systems through precise annotation.

Directed the final QA review and annotation of 2D hand and body keypoints for driver and passenger analysis. Maintained rigorous standards for anatomical accuracy to support models in gesture recognition and in-cabin safety monitoring. Identified and corrected pose estimation errors to improve the reliability of human-computer interaction (HCI) models. • Ensured high fidelity in label placement for human pose estimation. • Supported HCI and in-cabin safety models through granular labeling. • Reduced error rates via systematic QA processes. • Enhanced gesture recognition systems through precise annotation.

2024

Clip Labeller (Video & 4D LiDAR) | QA Lead & Annotator

Scale AI3D SensorTracking

Audited the temporal and spatial accuracy of 4D LiDAR datasets to enable seamless vehicle tracking in complex video sequences. Executed high-precision bounding box placement on sequential key frames for predictive perception in autonomous vehicle systems. Validated the integrity of large-scale time-series datasets used in real-time navigation. • Quality checked temporal consistency in multi-dimensional sequences. • Supported the enhancement of perception algorithms with precise annotation. • Ensured operational data quality exceeding 99% accuracy. • Contributed to real-time, safety-critical system datasets.

Audited the temporal and spatial accuracy of 4D LiDAR datasets to enable seamless vehicle tracking in complex video sequences. Executed high-precision bounding box placement on sequential key frames for predictive perception in autonomous vehicle systems. Validated the integrity of large-scale time-series datasets used in real-time navigation. • Quality checked temporal consistency in multi-dimensional sequences. • Supported the enhancement of perception algorithms with precise annotation. • Ensured operational data quality exceeding 99% accuracy. • Contributed to real-time, safety-critical system datasets.

2024

3D Spatial Mapping (Parknet & Scene Labelling) | Senior Annotator

CVAT3D SensorSegmentation

Performed semantic labeling within 3D LiDAR point clouds, classifying pedestrians, vehicles, and emergency trucks accurately. Labeled and defined park zones, infrastructure boundaries, and applied 3D bounding boxes to optimize tracking and parking operations. Supported robust collision avoidance by implementing granular human detection. • Enhanced self-driving models with precise spatial mapping. • Contributed semantic labels for perception and planning systems. • Oversaw object classification within large point cloud datasets. • Maintained quality and accuracy at scale for autonomous vehicle data.

Performed semantic labeling within 3D LiDAR point clouds, classifying pedestrians, vehicles, and emergency trucks accurately. Labeled and defined park zones, infrastructure boundaries, and applied 3D bounding boxes to optimize tracking and parking operations. Supported robust collision avoidance by implementing granular human detection. • Enhanced self-driving models with precise spatial mapping. • Contributed semantic labels for perception and planning systems. • Oversaw object classification within large point cloud datasets. • Maintained quality and accuracy at scale for autonomous vehicle data.

2022 - 2024

bonding boxes for automotive vehicles

Don T DiscloseImageBounding Box

bonding boxes for autonomous vehicle trucks and persons

bonding boxes for autonomous vehicle trucks and persons

2023 - 2023

Poles (Text Annotation & OCR) | QA Specialist

Label StudioTextEntity Ner Classification

Oversaw the QA for OCR-labeled text ensuring 100% legibility and correct class assignment for urban regulatory and navigation signage. Annotated text regions and created bounding boxes to support infrastructure mapping in autonomous systems. Enhanced dataset quality by validating the metadata of environmental labels. • Ensured accuracy and completeness in OCR results. • Contributed to infrastructure and street scene understanding. • Validated and annotated urban landmark data. • Maintained meticulous standards in text annotation processes.

Oversaw the QA for OCR-labeled text ensuring 100% legibility and correct class assignment for urban regulatory and navigation signage. Annotated text regions and created bounding boxes to support infrastructure mapping in autonomous systems. Enhanced dataset quality by validating the metadata of environmental labels. • Ensured accuracy and completeness in OCR results. • Contributed to infrastructure and street scene understanding. • Validated and annotated urban landmark data. • Maintained meticulous standards in text annotation processes.

2022 - 2022

Education

Work History