English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep learning vision systems are widely deployed across applications wherereliability is critical. However, even today's best models can fail torecognize an object when its pose, lighting, or background varies. Whileexisting benchmarks surface examples challenging for models, they do notexplain why such mistakes arise. To address this need, we introduce ImageNet-X,a set of sixteen human annotations of factors such as pose, background, orlighting the entire ImageNet-1k validation set as well as a random subset of12k training images. Equipped with ImageNet-X, we investigate 2,200 currentrecognition models and study the types of mistakes as a function of model's (1)architecture, e.g. transformer vs. convolutional, (2) learning paradigm, e.g.supervised vs. self-supervised, and (3) training procedures, e.g., dataaugmentation. Regardless of these choices, we find models have consistentfailure modes across ImageNet-X categories. We also find that while dataaugmentation can improve robustness to certain factors, they induce spill-overeffects to other factors. For example, strong random cropping hurts robustnesson smaller objects. Together, these insights suggest to advance the robustnessof modern vision models, future research should focus on collecting additionaldata and understanding data augmentation schemes. Along with these insights, werelease a toolkit based on ImageNet-X to spur further study into the mistakesimage recognition systems make.

ImageNet-X: Understanding Model Mistakes with Factor of Variation Annotations