Hybrid Deep Feature Fusion of 2D CNN and 3D CNN for Vestibule Segmentation from CT Images

Figure 7

The visualization comparison results of IEBL-DataSet. (a) The ground truth. (b) The segmented results of our proposed method. (c–e) The segmentation results obtained by using the three comparison methods, respectively.

(a) Ground truth

(b) Ours

(c) 2D CNN

(d) DeeplabV3+

(e) 3D DSD