Faces

eSciDoc Solutions PubMan: Overview · Functionalities Interfaces · Support Faces: Overview · Functionalities Scope · Support ViRR: Overview · Functionalities Scope · Support imeji Digitization Lifecycle edit

Faces is a collaboration of the MPDL with the Max Planck Institute for Human Development.

What is Faces

Overview

FACES is a lifespan database of adult emotional facial stimuli. It contains a set of images of naturalistic faces of 171 persons, separated in three different age groups (young, middle-aged and old) displaying each of 6 facial expressions: neutrality, sadness, disgust, fear, anger, and happiness.

For more information, please visit the FACES database homepage.

Details

Scope

• Developed between 2005-2007
• Currently 171 persons (men and women)
• 2 x 6 emotions
• 2052 pictures overall
• Available to defined user-group only as a java-application to be ordered by request on CD

Attributes

• ID, age group, age (at the photo shooting), gender, emotion, picture group

Emotions

• Neutrality, sadness, disgust, fear, anger, happiness

Age Groups

• Young (58 persons), middle-aged (56 persons), old (57 persons)

Picture Groups

• a and b (only as double check, no quality criteria)

Formats

• JPEG images
• Attribute export as CSV

Image Resolutions

• 819 x 1024
• 2835 x 3543
• Automatically generated thumbnails

What should Faces become

Faces should become an online open source solution, but with restricted access to the data because of the restricted usage rights for the photos.

Aim

The main focus of the project - next to the integration of Faces in the eSciDoc architecture - would be to add new attributes to single pictures in form of standardized metadata (e.g METS) and to implement a user management for different usage rights. Therefore, the application has to be flexible to integrate further data like the outcome of the norm study.

Functional Requirements

Browsing

• Browse photo collection (by five attributes)
• View photos and their attributes

Searching

• Select photos according to their attributes (advanced search)
• Manually refine selection

Export

• Export and save result set including attribute data
• The export should be also in CSV

Data Enrichment

• Adding of new attributes to images, also from external data via ingestion or one by one (new attributes have to be introduced to all images, also if the values are not available for all of them)
• Annotations of data sets (e.g blocking of a data set)
• Maybe integration of administrative data (e.g. birthday, contact details) in an administrative view
• Publications based on the pictures should be integrated in the application
• Currently, there is no need for adding further pictures
• No possibility for changing the pictures

User Management

• Visibility of the data only for registered users
• Online application form for an account
• Two different user roles: Reader and Editor

Creation of Subsets

• The creation and storage of subsets from the collection for different user groups should be supported

Metadata

• Development of a descriptive and a technical metadata set on collection and item level (maybe with a differentiation between individual-related and picture-related metadata)

Quality Assurance

• Currently, no quality assurance is necessary, because only a hand of people will get the Editor rights
• If later on, the group of Editors will be expanded to external scientists, the external data should be marked adequate (incl. a description of the used research method)

PID

• No PID (persistent identifier) for the individual photos is necessary, because the photos are not allowed to be cited, only the whole application, which therefore needs a PID

Technical Requirements

• Separation of the available metadata from the pictures (currently, the attributes are stored in the name of the pictures and the technical metadata directly in the JPEGs)

Corporate Design Requirements

The application has to be branded with the following:

• Name and logo of the institute
• Field of research or project level

Normstudy

Currently, the MPI is working on a normstudy, for comparing their expert opinion about how an emotion have to look like with the opinion of layman. The results of this study, about 130 grouped values per picture, should be integrated in FACES. Approximately, this will not start before the end of 2008.

This results are no further metadata for the pictures, but data. This data have to be connected to the pictures, but will be stored as separate items.

Meetings

19.06.07 (Berlin): Kick off Meeting

08./09. oder 10.10.07 (Berlin): Follow up Meeting

Max Planck Digital Library

Identifier

Image File Format

Comma-Separated Values

Enhanced Scientific Documentation

Persistent Identifer or Identification

Max-Planck-Institut