International audienceThis work investigates the use of deep fully convolutional neural networks (DFCNN) for pixel-wise scene labeling of Earth Observation images. Especially, we train a variant of the SegNet architecture on remote sensing data over an urban area and study different strategies for performing accurate semantic segmentation. Our contributions are the following : 1) we transfer efficiently a DFCNN from generic everyday images to remote sensing images ; 2) we perform data fusion from heterogeneous sensors (optical and laser) using residual correction. Our framework improves state-of-the-art accuracy on the ISPRS Vaihingen 2D Semantic Labeling dataset.Ce travail porte sur l'utilisation des réseaux de neu-rones convolutifs profon...
International audienceIn this work, we investigate various methods to deal with semantic labeling of...
In this paper, a novel convolutional neural network (CNN)-based architecture, named fine segmentatio...
International audienceSemantic segmentation is an essential part of deep learning. In recent years, ...
International audienceThis work investigates the use of deep fully convolutional neural networks (DF...
International audienceThis work investigates the use of deep fully convolutional neural networks (DF...
National audienceIn this paper, a novel method to tackle semantic segmentation of very high resoluti...
International audienceThis paper presents a novel semantic segmentation method of very high resoluti...
Semantic segmentation is a fundamental task in remote sensing image interpretation, which aims to as...
International audienceDeep learning (DL) is currently the dominant approach to image classification ...
We propose in this thesis to build up a collaboration between a deep neural network and a human in t...
Deep learning (DL) is currently the dominant approach to image classification and segmentation, but ...
This is the final version. Available from SPIE via the DOI in this recordSemantic segmentation is on...
International audienceThis paper explores different aspects of semantic segmentation of remote sensi...
Deep learning architectures have received much attention in recent years demonstrating state-of-the-...
International audienceIn this work, we investigate various methods to deal with semantic labeling of...
In this paper, a novel convolutional neural network (CNN)-based architecture, named fine segmentatio...
International audienceSemantic segmentation is an essential part of deep learning. In recent years, ...
International audienceThis work investigates the use of deep fully convolutional neural networks (DF...
International audienceThis work investigates the use of deep fully convolutional neural networks (DF...
National audienceIn this paper, a novel method to tackle semantic segmentation of very high resoluti...
International audienceThis paper presents a novel semantic segmentation method of very high resoluti...
Semantic segmentation is a fundamental task in remote sensing image interpretation, which aims to as...
International audienceDeep learning (DL) is currently the dominant approach to image classification ...
We propose in this thesis to build up a collaboration between a deep neural network and a human in t...
Deep learning (DL) is currently the dominant approach to image classification and segmentation, but ...
This is the final version. Available from SPIE via the DOI in this recordSemantic segmentation is on...
International audienceThis paper explores different aspects of semantic segmentation of remote sensi...
Deep learning architectures have received much attention in recent years demonstrating state-of-the-...
International audienceIn this work, we investigate various methods to deal with semantic labeling of...
In this paper, a novel convolutional neural network (CNN)-based architecture, named fine segmentatio...
International audienceSemantic segmentation is an essential part of deep learning. In recent years, ...