Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Pykognition

Schmøkel, Rasmus and Bossetta, Michael LU (2020)
Abstract
Python wrapper for AWS Rekognition API

Pykognition is a Python wrapper for the Amazon Web Service (AWS) Rekognition API, which provides industry-grade face and emotion detection. For facial detection, the algorithm provides a score for the predicted probability that the image includes a face (or multiple faces). Each face is categorized in a FaceDetail object, which carries a host of metadata such as predicted age, gender, and the emotion predicted to be displayed by each face.

The emotion classifications provided by the algorithm are: Happy, Sad, Angry, Confused, Disgusted, Surprised, Calm, Fear, and Unknown. Each emotion classification is accompanied by a confidence score ranging up to 99.9%.

Pykognition... (More)
Python wrapper for AWS Rekognition API

Pykognition is a Python wrapper for the Amazon Web Service (AWS) Rekognition API, which provides industry-grade face and emotion detection. For facial detection, the algorithm provides a score for the predicted probability that the image includes a face (or multiple faces). Each face is categorized in a FaceDetail object, which carries a host of metadata such as predicted age, gender, and the emotion predicted to be displayed by each face.

The emotion classifications provided by the algorithm are: Happy, Sad, Angry, Confused, Disgusted, Surprised, Calm, Fear, and Unknown. Each emotion classification is accompanied by a confidence score ranging up to 99.9%.

Pykognition simplifies the process of classifying images with the Rekognition API. Once the researcher establishes an AWS account, they only need to insert their access tokens and an input path where the images are stored. The ‘ifa’ function (short for Image Face Analysis) sends images for classification to the Rekognition API and returns both the classification and metadata.

Prerequisites
Python 3 (Is tested on 3.7 but might work on other versions)
Access to AWS Rekogntion API
(Less)
Please use this url to cite or link to this publication:
author
and
organization
publishing date
type
Non-textual form
publication status
published
subject
language
English
LU publication?
yes
id
9a4318e7-ebf4-4e88-8084-b75fe0e86e3e
alternative location
https://github.com/schmokel/Pykognition
date added to LUP
2022-01-04 10:13:07
date last changed
2022-01-17 11:26:12
@misc{9a4318e7-ebf4-4e88-8084-b75fe0e86e3e,
  abstract     = {{Python wrapper for AWS Rekognition API<br/><br/>Pykognition is a Python wrapper for the Amazon Web Service (AWS) Rekognition API, which provides industry-grade face and emotion detection. For facial detection, the algorithm provides a score for the predicted probability that the image includes a face (or multiple faces). Each face is categorized in a FaceDetail object, which carries a host of metadata such as predicted age, gender, and the emotion predicted to be displayed by each face.<br/><br/>The emotion classifications provided by the algorithm are: Happy, Sad, Angry, Confused, Disgusted, Surprised, Calm, Fear, and Unknown. Each emotion classification is accompanied by a confidence score ranging up to 99.9%.<br/><br/>Pykognition simplifies the process of classifying images with the Rekognition API. Once the researcher establishes an AWS account, they only need to insert their access tokens and an input path where the images are stored. The ‘ifa’ function (short for Image Face Analysis) sends images for classification to the Rekognition API and returns both the classification and metadata.<br/><br/>Prerequisites<br/>Python 3 (Is tested on 3.7 but might work on other versions)<br/>Access to AWS Rekogntion API<br/>}},
  author       = {{Schmøkel, Rasmus and Bossetta, Michael}},
  language     = {{eng}},
  title        = {{Pykognition}},
  url          = {{https://github.com/schmokel/Pykognition}},
  year         = {{2020}},
}