Classification robustness to common optical aberrations
This work addresses robustness issues in safety-critical computer vision applications by focusing on realistic optical degradations, though it is incremental as it builds on existing data augmentation and benchmark practices.
The paper tackles the problem of deep neural networks' robustness to realistic optical aberrations by proposing OpticsBench, a benchmark for evaluating such robustness, and shows that performance varies significantly compared to simpler blur models. It also introduces OpticsAugment, a data augmentation method using optical kernels, which improves robustness by 21.7% on OpticsBench and 6.8% on common corruptions.
Computer vision using deep neural networks (DNNs) has brought about seminal changes in people's lives. Applications range from automotive, face recognition in the security industry, to industrial process monitoring. In some cases, DNNs infer even in safety-critical situations. Therefore, for practical applications, DNNs have to behave in a robust way to disturbances such as noise, pixelation, or blur. Blur directly impacts the performance of DNNs, which are often approximated as a disk-shaped kernel to model defocus. However, optics suggests that there are different kernel shapes depending on wavelength and location caused by optical aberrations. In practice, as the optical quality of a lens decreases, such aberrations increase. This paper proposes OpticsBench, a benchmark for investigating robustness to realistic, practically relevant optical blur effects. Each corruption represents an optical aberration (coma, astigmatism, spherical, trefoil) derived from Zernike Polynomials. Experiments on ImageNet show that for a variety of different pre-trained DNNs, the performance varies strongly compared to disk-shaped kernels, indicating the necessity of considering realistic image degradations. In addition, we show on ImageNet-100 with OpticsAugment that robustness can be increased by using optical kernels as data augmentation. Compared to a conventionally trained ResNeXt50, training with OpticsAugment achieves an average performance gain of 21.7% points on OpticsBench and 6.8% points on 2D common corruptions.