Skip to content

A new computer vision algorithm for the recognition of AUs typically seen in most applications, their intensities, and a large number (23) of basic and compound emotion categories across databases. Additionally, images are annotated semantically with 421 emotion keywords.

Notifications You must be signed in to change notification settings

SaraSherinThomas/Emotionet

Repository files navigation

Emotionet

An accurate, real-time algorithm for the automatic annotation of a million facial expressions in the wild

A new computer vision al- gorithm for the recognition of AUs typically seen in most applications, their intensities, and a large number (23) of basic and compound emotion categories across databases. Additionally, images are annotated semantically with 421 emotion keywords.

Link to dataset : http://www.pitt.edu/~emotion/downloads.html

About

A new computer vision algorithm for the recognition of AUs typically seen in most applications, their intensities, and a large number (23) of basic and compound emotion categories across databases. Additionally, images are annotated semantically with 421 emotion keywords.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages