PROJECT TITLE :

Objectness to assist salient object detection

ABSTRACT:

When coping with salient object that contains many regions with different appearances, salient object detection can be a difficult task as usually solely elements of the salient object are highlighted and consistency between the salient regions is poor. This study tackles this downside by introducing objectness to help the salient object detection. Instead of treating objectness in the identical manner as other low-level cues (e.g. uniqueness, location etc.) for the determination of regional saliency values, the authors emphasise that objectness should additionally play a vital role in tuning the consistency between salient regions. The authors integrate objectness, uniqueness and centre bias to seek out potential salient regions and then enforce consistency between these regions using a full-connected Gaussian Markov random field with the weights determined by the objectness score. Experimental results on public benchmark datasets indicate that the authors' methodology performs well on many pictures which can not be well detected historically.