A Comprehensive Study of Image Classification Model Sensitivity to Foregrounds, Backgrounds, and Visual Attributes. (arXiv:2201.10766v1 [cs.CV])

Web: http://arxiv.org/abs/2201.10766

While datasets with single-label supervision have propelled rapid advances in
image classification, additional annotations are necessary in order to
quantitatively assess how models make predictions. To this end, for a subset of
ImageNet samples, we collect segmentation masks for the entire object and $18$
informative attributes. We call this dataset RIVAL10 (RIch Visual Attributes
with Localization), consisting of roughly $26k$ instances over $10$ classes.
Using RIVAL10, we evaluate the sensitivity of a broad set of models to noise
corruptions in …

arxiv classification cv model study