Research Article
Scene Understanding Based on High-Order Potentials and Generative Adversarial Networks
Figure 5
Comparison of different semantic segmentation algorithms on PASCAL VOC 2012: (a) original image; (b) FCN-8S; (c) DeepLab; (d) our method; (e) ground truth.
(a) |
(b) |
(c) |
(d) |
(e) |