Composite Statistical Inference for Semantic Segmentation

In this paper we present an inference procedure for the semantic segmentation of images. Different from many CRF approaches that rely on dependencies modeled with unary and pairwise pixel or superpixel potentials, our method is entirely based on estimates of the overlap between each of a set of mid-level object segmentation proposals and the objects present in the image. We define continuous latent variables on superpixels obtained by multiple intersections of segments, then output the optimal segments from the in-ferred superpixel statistics. The algorithm is capable of recombine and refine initial mid-level proposals, as well as handle multiple interacting objects, even from the same class, all in a consistent joint inference framework by maxi-mizing the composite likelihood of the underlying statistical model using an EM algorithm. In the PASCAL VOC seg-mentation challenge, the proposed approach obtains high accuracy and successfully handles images of complex ob-ject interactions. 1