Saved in:
Bibliographic Details
Main Authors: Safaei, Danial, Khastgir, Siddartha, Alirezaei, Mohsen, Ploeg, Jeroen, Tong, Son, Cheng, Chih-Hong, Zhao, Xingyu
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.16468
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914284905168896
author Safaei, Danial
Khastgir, Siddartha
Alirezaei, Mohsen
Ploeg, Jeroen
Tong, Son
Cheng, Chih-Hong
Zhao, Xingyu
author_facet Safaei, Danial
Khastgir, Siddartha
Alirezaei, Mohsen
Ploeg, Jeroen
Tong, Son
Cheng, Chih-Hong
Zhao, Xingyu
contents Virtual testing using synthetic data has become a cornerstone of autonomous vehicle (AV) safety assurance. Despite progress in improving visual realism through advanced simulators and generative AI, recent studies reveal that pixel-level fidelity alone does not ensure reliable transfer from simulation to the real world. What truly matters is whether the system-under-test (SUT) bases its decisions on consistent decision evidence in both real and simulated environments, not just whether images "look real" to humans. To this end this paper proposes a behavior-grounded fidelity measure by introducing Decisive Feature Fidelity (DFF), a new SUT-specific metric that extends the existing fidelity spectrum to capture mechanism parity, that is, agreement in the model-specific decisive evidence that drives the SUT's decisions across domains. DFF leverages explainable-AI methods to identify and compare the decisive features driving the SUT's outputs for matched real-synthetic pairs. We further propose estimators based on counterfactual explanations, along with a DFF-guided calibration scheme to enhance simulator fidelity. Experiments on 2126 matched KITTI-VirtualKITTI2 pairs demonstrate that DFF reveals discrepancies overlooked by conventional output-value fidelity. Furthermore, results show that DFF-guided calibration improves decisive-feature and input-level fidelity without sacrificing output value fidelity across diverse SUTs.
format Preprint
id arxiv_https___arxiv_org_abs_2512_16468
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Quantifying Fidelity: A Decisive Feature Approach to Comparing Synthetic and Real Imagery
Safaei, Danial
Khastgir, Siddartha
Alirezaei, Mohsen
Ploeg, Jeroen
Tong, Son
Cheng, Chih-Hong
Zhao, Xingyu
Artificial Intelligence
Virtual testing using synthetic data has become a cornerstone of autonomous vehicle (AV) safety assurance. Despite progress in improving visual realism through advanced simulators and generative AI, recent studies reveal that pixel-level fidelity alone does not ensure reliable transfer from simulation to the real world. What truly matters is whether the system-under-test (SUT) bases its decisions on consistent decision evidence in both real and simulated environments, not just whether images "look real" to humans. To this end this paper proposes a behavior-grounded fidelity measure by introducing Decisive Feature Fidelity (DFF), a new SUT-specific metric that extends the existing fidelity spectrum to capture mechanism parity, that is, agreement in the model-specific decisive evidence that drives the SUT's decisions across domains. DFF leverages explainable-AI methods to identify and compare the decisive features driving the SUT's outputs for matched real-synthetic pairs. We further propose estimators based on counterfactual explanations, along with a DFF-guided calibration scheme to enhance simulator fidelity. Experiments on 2126 matched KITTI-VirtualKITTI2 pairs demonstrate that DFF reveals discrepancies overlooked by conventional output-value fidelity. Furthermore, results show that DFF-guided calibration improves decisive-feature and input-level fidelity without sacrificing output value fidelity across diverse SUTs.
title Quantifying Fidelity: A Decisive Feature Approach to Comparing Synthetic and Real Imagery
topic Artificial Intelligence
url https://arxiv.org/abs/2512.16468