Making the Discrete Continuous: Synthetic RAW Augmentations for Fine-Grained Evaluation of Person Detection Performance in Low Light

2026-05-21

AI Summary

This work addresses the scarcity and uneven distribution of real-world pedestrian detection data under low-light conditions, which hinders fine-grained performance evaluation. To bridge this gap, the study introduces— for the first time—a RAW image synthesis method grounded in the physical noise model of camera sensors, enabling continuous expansion of the low-light input space to generate high-fidelity synthetic samples. This approach substantially enhances dataset coverage for benchmarking and demonstrates strong performance alignment between synthetic and real low-light data across multiple state-of-the-art object detectors. By effectively closing the evaluation gap, the proposed framework establishes a reliable and scalable paradigm for assessing pedestrian detection in low-light scenarios.

Abstract

Real-world deployment of AI vision models is both fueled and limited by the data available for training and testing. Real datasets are sparse and uneven: long-tailed or unbalanced distributions hinder generalization, and the low number of samples in low density regions makes it hard to run evaluations. Synthetic data can fill these gaps, providing us with a way to sample the input space more continuously and improve data coverage for benchmarks. Focusing on the autonomous driving safety-critical case of pedestrian detection in the dark, we show how synthetic low-light samples can be used to better characterize the performance of a state-of-the-art object detection model as a function of the scene illumination. We use a synthetic RAW image augmentation technique to generate low-light samples that match the noise model of the camera sensor. Performance metrics on real and synthetic low-light data are similar, indicating that the AI model finds it hard to distinguish between them.

Problem

• low light
• person detection
• synthetic data
• performance evaluation
• data scarcity

Innovation

• synthetic RAW augmentation
• low-light pedestrian detection
• sensor noise modeling
• fine-grained evaluation
• continuous data sampling