Abstract

We present an RGBD video segmentation method that takes advantage of depth data and can extract multiple foregrounds in the scene. This video segmentation is addressed as an object proposal selection problem formulated in a fullyconnected graph where a flexible number of foregrounds may be chosen. In the graph, each node represents a proposal, and the edges model intra-frame and inter-frame constraints on the solution.

The proposals are generated based on an RGBD video saliency map in which depth-based features are utilized to enhance identification of foregrounds. Experiments show that the proposed multiple foreground segmentation method outperforms related techniques, and the depth cue serves as a helpful complement to RGB features. Moreover, our method provides performance comparable to state-of-the-art RGB video segmentation techniques on regular RGB videos with estimated depth maps.