The purpose of the Grand Challenge Saliency4ASD is to drive efforts of visual attention modeling community towards a healthcare societal challenge. Gaze features related to saccades and fixations have demonstrated their usefulness in the identification of mental states, cognitive processes and neuropathologies (Tseng et al., 2013; Itti, 2015), notably for people with ASD (Autism Spectrum Disorder).
With this Grand challenge, it is expected, from datasets released by the organizers:
- track 1) to develop models (saliency prediction or saccadic models) that fit gaze behavior of people with ASD;
- track 2) to propose models able to classify ASD and normal viewers using gaze data.
The first goal is particularly useful to develop ad-hoc CHIs (Computer Human Interfaces) that are adapted for people with ASD. In addition to early stage detection of ASD, the second goal could be relevant to monitor the efficiency of any remediation protocol.
Participants are free to submit computational models in the both categories (Each of the submissions will be evaluated with the respective collected ground-truth by the organizers).
For more information do not hesitate to send an email to saliency4ASD@ls2n.fr
The organizing team:
Guangtao Zhai and Zhaohui Che (Shanghai Jao Tong University, China)
Jesús Gutiérrez and Patrick Le Callet (Université de Nantes, France)