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

TL;DR

This paper introduces a synthetic RAW augmentation method to generate low-light images for better evaluation of pedestrian detection models in dark conditions, addressing data scarcity and imbalance.

Contribution

The authors propose a synthetic RAW image augmentation technique that creates realistic low-light samples to improve performance evaluation in low-light scenarios.

Findings

Synthetic low-light samples match real data performance metrics.

AI models struggle to distinguish between real and synthetic low-light images.

Synthetic augmentation enhances data coverage for low-light evaluation.

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.