Skip to main content
. Author manuscript; available in PMC: 2014 May 1.
Published in final edited form as: IEEE Trans Image Process. 2013 Jun 21;22(10):3866–3878. doi: 10.1109/TIP.2013.2263147

Fig. 4.

Fig. 4

Zebra image. (a) The initial boxes for target and background (b) Error-difference image (c) Result by Lazysnap (d) Result by Vector valued CV (e) Result by GrabCut (f) STAC (g) The seeds (green: target bounding box; blue: background) for the grabCut in order to obtain the result in (h).