English

3D Densely Convolutional Networks for Volumetric Segmentation

Computer Vision and Pattern Recognition 2017-09-15 v2

Abstract

In the isointense stage, the accurate volumetric image segmentation is a challenging task due to the low contrast between tissues. In this paper, we propose a novel very deep network architecture based on a densely convolutional network for volumetric brain segmentation. The proposed network architecture provides a dense connection between layers that aims to improve the information flow in the network. By concatenating features map of fine and coarse dense blocks, it allows capturing multi-scale contextual information. Experimental results demonstrate significant advantages of the proposed method over existing methods, in terms of both segmentation accuracy and parameter efficiency in MICCAI grand challenge on 6-month infant brain MRI segmentation.

Keywords

Cite

@article{arxiv.1709.03199,
  title  = {3D Densely Convolutional Networks for Volumetric Segmentation},
  author = {Toan Duc Bui and Jitae Shin and Taesup Moon},
  journal= {arXiv preprint arXiv:1709.03199},
  year   = {2017}
}

Comments

7 pages

R2 v1 2026-06-22T21:38:32.629Z