Residual Conv-Deconv Grid Network for Semantic Segmentation

Damien Fourure (1, 2) , Rémi Emonet (1) , Elisa Fromont (1) , Damien Muselet (1) , Alain Tremeau (1) , Christian Wolf (2)

1 LHC - Laboratoire Hubert Curien
2 imagine - Extraction de Caractéristiques et Identification

This paper presents GridNet, a new Convolutional Neural Network (CNN) architecture for semantic image segmentation (full scene labelling). Classical neural networks are implemented as one stream from the input to the output with subsampling operators applied in the stream in order to reduce the feature maps size and to increase the receptive field for the final prediction. However, for semantic image segmentation, where the task consists in providing a semantic class to each pixel of an image, feature maps reduction is harmful because it leads to a resolution loss in the output prediction. To tackle this problem, our GridNet follows a grid pattern allowing multiple interconnected streams to work at different resolutions. We show that our network generalizes many well known networks such as conv-deconv, residual or U-Net networks. GridNet is trained from scratch and achieves competitive results on the Cityscapes dataset.

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]; Intelligence artificielle [cs.AI]; Réseau de neurones [cs.NE]

Damien Fourure, Rémi Emonet, Elisa Fromont, Damien Muselet, Alain Tremeau, et al.. Residual Conv-Deconv Grid Network for Semantic Segmentation. BMVC 2017, Sep 2017, Londre, United Kingdom. ⟨hal-01567725⟩