Image Co-segmentation Using Maximum Common Subgraph Matching and Region Co-growing
We propose a computationally efficient graph based image co-segmentation algorithm where we extract objects with similar features from an image pair or a set of images. First we build a region adjacency graph (RAG) for each image by representing image sup
- PDF / 2,094,003 Bytes
- 17 Pages / 439.37 x 666.142 pts Page_size
- 37 Downloads / 176 Views
Abstract. We propose a computationally efficient graph based image co-segmentation algorithm where we extract objects with similar features from an image pair or a set of images. First we build a region adjacency graph (RAG) for each image by representing image superpixels as nodes. Then we compute the maximum common subgraph (MCS) between the RAGs using the minimum vertex cover of a product graph obtained from the RAG. Next using MCS outputs as the seeds, we iteratively co-grow the matched regions obtained from the MCS in each of the constituent images by using a weighted measure of inter-image feature similarities among the already matched regions and their neighbors that have not been matched yet. Upon convergence, we obtain the co-segmented objects. The MCS based algorithm allows multiple, similar objects to be co-segmented and the region co-growing stage helps to extract different sized, similar objects. Superiority of the proposed method is demonstrated by processing images containing different sized objects and multiple objects. Keywords: Maximum common subgraph
1
· Region co-growing
Introduction
Co-segmentation is the problem of segmenting objects with similar features from more than one image (see Fig. 1) or from multiple frames in a video. The objects of common interest in multiple images are detected as co-segmented objects [1], [2], [3]. Image foreground segmentation without supervision is a difficult problem. If an additional image containing a similar foreground is provided, both images can be segmented simultaneously with a higher accuracy using cosegmentation. Co-segmentation can also be used to detect objects of common Partial financial supports from Bharti Centre for Communication in IIT Bombay and K.N Bajaj Chair Professorship are gratefully acknowledged. Electronic supplementary material The online version of this chapter (doi:10. 1007/978-3-319-46466-4 44) contains supplementary material, which is available to authorized users. c Springer International Publishing AG 2016 B. Leibe et al. (Eds.): ECCV 2016, Part VI, LNCS 9910, pp. 736–752, 2016. DOI: 10.1007/978-3-319-46466-4 44
Image Co-segmentation
737
Fig. 1. Illustration of the co-segmentation problem. (a–e) Images retrieved by a child from the internet, when asked to provide pictures of a tiger, and (f–j) common object quite apparent from the given set of images (Color figure online) Superpixel I 1 Segmented Input Image 2 I Pair
RAG with node attributes
G1
1 Product M Graph and MVC Computation M2
RAG with 2 node attributes G MCS Computation
Region Growing M1,(t) M2,(t) Region Growing
M1∗ (Object1) M2∗ (Object2)
Region Co-growing
Fig. 2. Block diagram of the proposed co-segmentation algorithm. Input image pair I 1 and I 2 is represented as region adjacency graphs (RAGs) G1 and G2 that are used to obtain the maximum common subgraph (MCS) that gives the initial matched regions M1 and M2 in I 1 and I 2 . These are iteratively (index-(t)) co-grown to obtain the final matched regions M1∗ and M2∗ . In order to grow the region M1 in I 1 , t
Data Loading...
