SS-SFR: Synthetic Scenes Spatial Frequency Response on Virtual KITTI and Degraded Automotive Simulations for Object Detection
This work addresses the problem of evaluating image quality in automotive simulations for researchers and developers in computer vision, though it is incremental as it applies existing degradation methods to a specific dataset.
The study investigated the impact of Gaussian blur on image sharpness in Virtual KITTI simulations and its effect on object detection performance, finding that while sharpness degraded significantly from 0.245cy/px to 0.119cy/px, detection performance remained robust with minimal drops of 0.58% to 1.93% across three state-of-the-art models.
Automotive simulation can potentially compensate for a lack of training data in computer vision applications. However, there has been little to no image quality evaluation of automotive simulation and the impact of optical degradations on simulation is little explored. In this work, we investigate Virtual KITTI and the impact of applying variations of Gaussian blur on image sharpness. Furthermore, we consider object detection, a common computer vision application on three different state-of-the-art models, thus allowing us to characterize the relationship between object detection and sharpness. It was found that while image sharpness (MTF50) degrades from an average of 0.245cy/px to approximately 0.119cy/px; object detection performance stays largely robust within 0.58\%(Faster RCNN), 1.45\%(YOLOF) and 1.93\%(DETR) across all respective held-out test sets.