Making the Discrete Continuous: Synthetic RAW Augmentations for Fine-Grained Evaluation of Person Detection Performance in Low Light
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.
Related
- Quantum steering in networks: Measurement-device-independent detection, continuous variables, and practical Gaussian schemes
- Performance of a Gamma Detection Unit with Likelihood-Based Localization Method
- First evaluation of the tracking performance of the ARCADIA Fully Depleted MAPS with high-energy proton beam
- Parameter-Efficient Quantum NLP for Paraphrase Detection: Performance, Robustness, and Entanglement
- Performance Optimization of Short Reach Optical Interconnects based on Direct Detection
BibTeX
@misc{pais2026making,
title = {{Making the Discrete Continuous: Synthetic RAW Augmentations for Fine-Grained Evaluation of Person Detection Performance in Low Light}},
author = {Valeria Pais and Malena Mendilaharzu and Daniele Faccio and Luis Oala and Christoph Clausen and Bruno Sanguinetti},
howpublished = {arXiv},
year = {2026},
month = aug,
doi = {10.48550/arxiv.2605.22455},
eprint = {2605.22455},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.22455},
note = {CC BY-SA 4.0},
}
RIS
TY - UNPB
TI - Making the Discrete Continuous: Synthetic RAW Augmentations for Fine-Grained Evaluation of Person Detection Performance in Low Light
AU - Valeria Pais
AU - Malena Mendilaharzu
AU - Daniele Faccio
AU - Luis Oala
AU - Christoph Clausen
AU - Bruno Sanguinetti
PY - 2026
DA - 2026/08/31
PB - arXiv
DO - 10.48550/arxiv.2605.22455
UR - https://arxiv.org/abs/2605.22455
AB - 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.
ER -