Skip to main content
Fig. 4 | IPSJ Transactions on Computer Vision and Applications

Fig. 4

From: Pedestrian segmentation based on a spatio-temporally consistent graph-cut with optimal transport

Fig. 4

Example of the edge cost function. a Input image of the frame t. b Clipping around the i-th pixel. Edge probability \(p^{t}_{\text {edge}} = 0.9\) on the left side (as represented by red) and \(p^{t}_{\text {edge}} = 0.9\) in the middle and on the right side (as represented by blue). c Edge cost of assigning the label l1 to the i-th pixel cedge(i,l1)=−0.1 while cedge(i,l2)=−0.9; therefore, l2 is more likely to be assigned to the i-th pixel

Back to article page