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

Faces form the basis for a rich variety of judgments in humans, yet theunderlying features remain poorly understood. Although fine-graineddistinctions within a race might more strongly constrain possible facialfeatures used by humans than in case of coarse categories such as race orgender, such fine grained distinctions are relatively less studied.Fine-grained race classification is also interesting because even humans maynot be perfectly accurate on these tasks. This allows us to compare errors madeby humans and machines, in contrast to standard object detection tasks wherehuman performance is nearly perfect. We have developed a novel face database ofclose to 1650 diverse Indian faces labeled for fine-grained race (South vsNorth India) as well as for age, weight, height and gender. We then asked closeto 130 human subjects who were instructed to categorize each face as belongingtoa Northern or Southern state in India. We then compared human performance onthis task with that of computational models trained on the ground-truth labels.Our main results are as follows: (1) Humans are highly consistent (averageaccuracy: 63.6%), with some faces being consistently classified with > 90%accuracy and others consistently misclassified with < 30% accuracy; (2) Modelstrained on ground-truth labels showed slightly worse performance (averageaccuracy: 62%) but showed higher accuracy (72.2%) on faces classified with >80% accuracy by humans. This was true for models trained on simple spatial andintensity measurements extracted from faces as well as deep neural networkstrained on race or gender classification; (3) Using overcomplete banks offeatures derived from each face part, we found that mouth shape was the singlelargest contributor towards fine-grained race classification, whereas distancesbetween face parts was the strongest predictor of gender.

Can you tell where in India I am from? Comparing humans and computers on fine-grained race face classification