This paper proposes a new deep convolutional neural network (DCNN) architecture that learns pixel embeddings, such that pairwise distances between the embeddings can be used to infer whether or not the pixels lie on the same region. That is, for any two pixels on the same object, the embeddings are trained to be similar; for any pair that straddles an object boundary, the embeddings are trained to be dissimilar. Experimental results show that when this embedding network is used in conjunction with a DCNN trained on semantic segmentation, there is a systematic improvement in per-pixel classification accuracy. Our contributions are integrated in the popular Caffe deep learning framework, and consist in straightforward modifications to convolution routines. As such, they can be exploited for any task involving convolution layers.
@article{arxiv.1511.04377,
title = {Learning Dense Convolutional Embeddings for Semantic Segmentation},
author = {Adam W. Harley and Konstantinos G. Derpanis and Iasonas Kokkinos},
journal= {arXiv preprint arXiv:1511.04377},
year = {2016}
}