The majority of commercial facial-recognition systems exhibit bias, according to a study from a federal agency released on Thursday, underscoring questions about a technology increasingly used by police departments and federal agencies to identify suspected criminals. The systems falsely identified African-American and Asian faces 10 times to times more than Caucasian faces, the National Institute of Standards and Technology reported on Thursday. Among a database of photos used by law enforcement agencies in the United States, the highest error rates came in identifying Native Americans, the study found. The technology also had more difficulty identifying women than men.
Many Facial-Recognition Systems Are Biased, Says U.S. Study
Multiple facial for a super cute girl | themadexperience.com
The perception of facial expressions and objects at a distance are entrenched psychological research venues, but their intersection is not. We were motivated to study them together because of their joint importance in the physical composition of popular movies—shots that show a larger image of a face typically have shorter durations than those in which the face is smaller. For static images, we explore the time it takes viewers to categorize the valence of different facial expressions as a function of their visual size. In two studies, we find that smaller faces take longer to categorize than those that are larger, and this pattern interacts with local background clutter. More clutter creates crowding and impedes the interpretation of expressions for more distant faces but not proximal ones. Filmmakers at least tacitly know this.
Facial expression, size, and clutter: Inferences from movie structure to emotion judgments and back
Search SpringerLink Search. This dataset contains a lot of images with almost each possible facial expression. Images are downloaded from web therefore these are not of fixed resolution.