In this poster, we propose to face the problem of event detec/on from single
images, by exploi/ng both background informaAon oNen containing revealing
contextual clues and details, which are salient for recognizing the event. Such
details are visual objects criAcal to understand the underlying event depicted in
the image and were recently defined in the literature as ”event-saliency”.
Adop/ng the MulAple-Instance Learning (MIL) paradigm we propose a
hierarchical approach analyzing first the en/re picture and then refining the
decision on the basis of the event-salient objects. Valida/on of the proposed
method is carried out on two benchmarking datasets and it demonstrates the
effec/veness of the proposed hierarchical approach to event discovery from single
images.