Motivated professional with 3.5 years of experience in data analysis and video annotation. Expertise in AI-driven automotive technologies with a strong focus on detail and accuracy. Proven ability to contribute to team success and drive continuous improvement in processes.
Scene Hazards
• Labeled potential driving hazards like construction zones, obstacles, and blind turns in complex road environments.
• Classified objects into subclasses such as Traffic Objects (e.g., cone, pole, barrier, barricade, barrel, triangle) and Hazard Objects (e.g., debris vehicle, debris natural, tire, human, animal).
• Supported detection and classification models with accurate annotation of unclassifiable and grouped hazard types.
• Contributed to enhancing safety algorithms in autonomous driving modules.
Scene Labelling
• Labeled obstacles that impact vehicle movement such as parked vehicles, pedestrians, and animals.
• Used predefined categories such as Vehicles (e.g., automobiles, buses, trailers), VRUs (e.g., persons, riders, strollers), Animals (on and off-road), and Protruding Objects (e.g., cargo, open doors).
• Focused on outlining geometric structures and determining the accurate bounding shapes for all detected obstacles.
• Helped in training machine learning models to better adapt to varied driving conditions.
Wait Light Condition
• Annotated traffic light states (green, yellow, red) and corresponding vehicle behavior in intersection scenarios.
• Evaluated attributes like blinking status, bulb visibility, and active bulb count to enhance signal recognition models.
• Supported datasets that improve decision-making logic in self-driving systems.