Instance-Level Safety-Aware Fidelity of Synthetic Data and Its Calibration

Abstract

Modeling and calibrating the fidelity of synthetic data is paramount in shaping the future of safe and reliable self-driving technology by offering a cost-effective and scalable alternative to real-world data collection. We focus on its role in safety-critical applications, introducing four types of instance-level fidelity that go beyond mere visual input characteristics. The aim is to ensure that applying testing on synthetic data can reveal real-world safety issues, and the absence of safety-critical issues when testing under synthetic data can provide a strong safety guarantee in real-world behavior. We suggest an optimization method to refine the synthetic data generator, reducing fidelity gaps identified by deep learning components. Experiments show this tuning enhances the correlation between safety-critical errors in synthetic and real data.

References in corpus (6)

• SSD: Single Shot MultiBox Detector
• Deep Visual Domain Adaptation: A Survey
• Benchmarking Detection Transfer Learning with Vision Transformers
• GAIA-1: A Generative World Model for Autonomous Driving
• A taxonomy for quality in simulation-based development and testing of automated driving systems
• Revisiting 3D Object Detection From an Egocentric Perspective