PROJECT TITLE :

Image Co-Saliency Detection and Co-Segmentation via Progressive Joint Optimization

ABSTRACT:

New computational models for simultaneous picture co-saliency detection and co-segmentation are presented here that simultaneously investigate the ideas of saliency and objectness in numerous images. Multiple visual signals, such as colour and texture, can be combined to better identify salient objects; however, the ideal recommendations are often region-dependent and the fusion process often results in blurred results. Co-segmentation can help to preserve object boundaries, although it may suffer in complicated scenarios.. By tackling an energy-minimization problem over the graph, we offer a strategy that simultaneously handles co-saliency detection and the problem of co-segmentation. When a region-wise adaptive saliency map and object segmentation are combined, relevant information is transferred between the two complimentary tasks. Salient objects can be recovered via object segmentation and the optimization iterations, which provide increasingly crisp saliency maps. The improved saliency prior allows these segmentations to be improved further. We compare our approach to the current state of the art using four publicly available benchmark data sets. For both co-saliency detection and co-segmentation, our strategy consistently delivers better results after extensive testing.