This is a lightweight and versatile robustness benchmark built upon the training set of ImageNet-1K. It contains an overall of 50,000 images, divided in 5 components, evenly distributed over 1,000 classes. It assesses the performance of a classification model in five aspects: accuracy on intrinsically difficult images (SuperHard, SH), images with partial information (PartialInfo, PI), robustness against low resolution (LowResolution, LR), adversarial attacks (AdversarialAttack, AA), and speckle noise (SpeckleNoise, SN).

Dataset Files

You must be an IEEE Dataport Subscriber to access these files. Subscribe now or login.

[1] Yiming Bian, "SPLAS", IEEE Dataport, 2024. [Online]. Available: http://dx.doi.org/10.21227/xag0-3668. Accessed: Feb. 18, 2025.
@data{xag0-3668-24,
doi = {10.21227/xag0-3668},
url = {http://dx.doi.org/10.21227/xag0-3668},
author = {Yiming Bian },
publisher = {IEEE Dataport},
title = {SPLAS},
year = {2024} }
TY - DATA
T1 - SPLAS
AU - Yiming Bian
PY - 2024
PB - IEEE Dataport
UR - 10.21227/xag0-3668
ER -
Yiming Bian. (2024). SPLAS. IEEE Dataport. http://dx.doi.org/10.21227/xag0-3668
Yiming Bian, 2024. SPLAS. Available at: http://dx.doi.org/10.21227/xag0-3668.
Yiming Bian. (2024). "SPLAS." Web.
1. Yiming Bian. SPLAS [Internet]. IEEE Dataport; 2024. Available from : http://dx.doi.org/10.21227/xag0-3668
Yiming Bian. "SPLAS." doi: 10.21227/xag0-3668