Call for Participation: Multimedia Information Processing for Personality & Social Network Analysis Challenge @ ICPR 2018

[Hugo Jair]

[Apologies for multiple postings]

***********************************************************************
1st CALL FOR PARTICIPATION
Multimedia Information Processing for Personality & Social Networks Analysis Challenge
2018 International Conference on Pattern Recognition
August 20-24, Beijing, China
Web site: http://chalearnlap.cvc.uab.es/challenge/27/description/
************************************************************************

Multimedia information processing is a fruitful research topic which addressed a hefty number of tasks, among them, those focusing on the analysis of human behavior. Although great advances have been made in the so-called Looking At People field, it is only recently that attention from this area is targeting problems that have to do with more complex behaviors. For instance, personality and social behaviors are just starting to be explored from a multimedia information processing perspective.

In this context, the ICPR 2018 Multimedia Information Processing for Personality & Social Networks Analysis challenge prospective participants on the analysis of non-obvious human behavior with two tasks:

***DivFusion***
Information fusion for social image retrieval and diversification. Diversification of image search results is now a hot research problem in multimedia. Search engines are now fostering techniques that allow for providing the user with a diverse representation of his search results, rather than providing redundant information, e.g. the same perspective of a monument, or location etc. The DivFusion task builds on the MediaEval Retrieving Diverse Social Images Tasks and challenges the participants to develop highly effective information fusion techniques. The participants will be provided with several query results, content descriptors and output of various existing diversification systems. They are to employ fusion strategies to refine the retrieval results thus to improve even more the diversification. The data consist of hundreds of Flickr image query results (>600 queries, both single- and multi- topic) and include: images (up to 300 images per query), social metadata, descriptors for visual, text, social information as well as deep learning features, expert annotations for image relevance and diversification (i.e. clustering of images according to the similarity of their content) and more than 180 diversification system outputs.


***HWxPI***
Estimating the personality traits of users from their handwritten texts and the corresponding transcripts (image and text modalities). The challenge comprises two phases: a development and a final phase. For the first phase, the participants should develop their systems using a set of development pairs of handwritten essays, including image and text from 418 subjects. Each subject has an associated class (either 1 or 0) corresponding to the presence of a high pole or a low pole of a specific personality trait. The traits correspond to the Big Five personality models used in psychology: Extraversion, Agreeableness, Conscientiousness, Emotional stability, and Openness to experience. Participants will have to develop a classifier to predict the pole of each trait by including both modalities (i.e. text and visual). For the final evaluation phase, an independent set of 293 unlabeled samples will be provided to the participants, who will have to provide predictions using the models trained on the development data.


***********************
Target communities
***********************
Both tracks lie at the frontier of research on Looking At People and multimedia information processing, and target communities such as (but are not limited to): information retrieval (text, vision, multimedia, social media, etc.), information fusion, machine learning, deep learning, data mining, natural language processing, image and text processing.

The challenges will run in the CodaLab platform (http://codalab.org/), and results will be presented at the IAPR ICPR 2018 conference in Beijing, China (http://www.icpr2018.org/). Participants obtaining the best results will be invited to submit a paper to a dedicated Special Issue organized with a top tier journal from the field (TBA).


******************************
Important dates (tentative)
******************************
- Competition starts (& release of the data): February 25, 2018
- Competition ends: April 24, 2018
- Challenge award ceremony: August 21, 2018 (@ ICPR 2018)


***********************
Overall coordination
***********************
Hugo Jair Escalante, Instituto Nacional de Astrofisica, Optica y Electronica, Mexico (hugo...@gmail.com)
Bogdan Ionescu, University Politehnica of Bucharest, Romania
Esaú Villatoro, Universidad Autonoma Metropolitana, campus Cuajimalpa, Mexico
Gabriela Ramírez, Universidad Autonoma Metropolitana, campus Cuajimalpa, Mexico
Sergio Escalera, Computer Vision Center & University of Barcelona, Barcelona, Spain
Martha Larson, Delft University of Technology, Netherlands
Henning Müller, University of Applied Sciences Western Switzerland (HES-SO), Switzerland