SegNetRes-CRF: A Deep Convolutional Encoder-Decoder Architecture for Semantic Image Segmentation

Semantic segmentation is an essential task in computer vision that aims to label each image pixel. Several of the actual best approaches in this context are based on deep neural networks. For example, SegNet is a deep encoderdecoder architecture approach whose results were disruptive because it is f...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2018 International Joint Conference on Neural Networks (IJCNN) S. 1 - 6
Hauptverfasser: de Oliveira Junior, Luiz Antonio, Medeiros, Heitor R., Macedo, David, Zanchettin, Cleber, Oliveira, Adriano L. I., Ludermir, Teresa
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.07.2018
Schlagworte:
Computer architecture
CRF
Decoding
Encoder-decoder architectures
Image segmentation
Neural networks
Residual networks
SegNet
Semantics
Task analysis
Training
U-Net
ISSN:2161-4407
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Semantic segmentation is an essential task in computer vision that aims to label each image pixel. Several of the actual best approaches in this context are based on deep neural networks. For example, SegNet is a deep encoderdecoder architecture approach whose results were disruptive because it is fast and performs well. However, this architecture fails to finedelineating the edges between the objects of interest in the image. We propose some modifications in the SegNet-Basic architecture by using a post-processing segmentation layer (using Conditional Random Fields) and by transferring high resolution features combined to the decoder network. The proposed method was evaluated in the dataset CamVid. Moreover, it was compared with important variants of SegNet and showed to be able to improve the overall accuracy of SegNet-Basic by up to 17.5%.
ISSN:2161-4407
DOI:10.1109/IJCNN.2018.8489376