You are here

ImageCLEFaware

Motivation

YDSYO app overview

Images constitute a large part of the content shared on social networks. Their disclosure is often related to a particular context and users are often unaware of the fact that, depending on their privacy status, images can be accessible to third parties and be used for purposes which were initially unforeseen. For instance, it is common practice for employers to search information about their future employees online. Another example of usage is that of automatic credit scoring based on online data. Most existing approaches which propose feedback about shared data focus on inferring user characteristics and their practical utility is rather limited.

We hypothesize that user feedback would be more efficient if conveyed through the real-life effects of data sharing. The objective of the task is to automatically score user photographic profiles in a series of situations with strong impact on her/his life. Four such situations were modeled this year and refer to searching for: (1) a bank loan, (2) an accommodation, (3) a job as waitress/waiter and (4) a job in IT. The inclusion of several situations is interesting in order to make it clear to the end users of the system that the same image will be interpreted differently depending on the context.

The final objective of the task is to encourage the development of efficient user feedback, such as the YDSYO Android app.
The current functioning is summarized in this video.

News

More information will be added soon!

Preliminary Schedule

    To be added soon.

Task description

Participants are required to provide automatic rankings of photographic user profiles in a series of real-life situations such as searching for a job, accommodation, insurance, bank loans, etc. The ranking will be based on an automatic analysis of profile images and the aggregation of individual results.

Data

A new data set will be proposed consisting of three main components: (i) user profiles with manual ratings (>=6,000 profiles) for each evaluated situation (4 situations) divided into train-validation-test subsets, (ii) visual class ratings (>3,500 classes per situation), and (iii) visual classes detection using a recent object detector such as EfficientDet.

Evaluation methodology

The evaluation will be performed using a classical correlation measure, such as Pearson’s Correlation Coefficient, between automatic user profile ranking and a manually obtained ground truth ranking obtained by crowdsourcing.

Participant registration

Please refer to the general ImageCLEF registration instructions.

Submission instructions

More information will be added soon!

Results

More information will be added soon!

CEUR Working Notes


More information will be added soon.

References

Van-Khoa Nguyen, Adrian Popescu, and Jerome Deshayes-Chossart. "Unveiling Real-Life Effects of Online Photo Sharing." IEEE WACV 2022.

Organizers

Organizers:

  • Jérôme Deshayes <jerome.deshayes-chossart(at)cea.fr>, CEA LIST, France
  • Adrian Popescu  <adrian.popescu(at)cea.fr>, CEA LIST, France
  • Bogdan Ionescu <bogdan.ionescu(at)upb.ro>, Politehnica University of Bucharest, Romania

Acknowledgements