抄録
H-006
Predicting Focus of Attention of Elderly Drivers
Onkar Krishna・Go Irie・Takahito Kawanishi・Kunio Kashino(NTT)・Kiyoharu Aizawa(The Univ. of Tokyo)
Predicting which part of a scene elderly people would pay attention could be useful for assisting elderly drivers. Cognitive studies suggest that elderly's vision gets impaired due to changes associated with aging. While many computational models for predicting focus of attention (FoA) have been developed, they are mostly focused on adult's vision, which may not work well for predicting elderly's. Is it possible to leverage the prediction results made by an FoA model of general adults to accurately predict elderly's FoA? We consider a novel problem of translating adult's FoA to elderly's and propose an approach based on deep image-to-image translation model. We test our model in the task of predicting what an elderly driver would pay attention to while car driving, and demonstrate that our model gives remarkable prediction accuracy compared to the baselines.