High-resolution image classification with convolutional networks

Emmanuel Maggiori 1 Yuliya Tarabalka 1 Guillaume Charpiat 2 Pierre Alliez 3, 1
1 TITANE - Geometric Modeling of 3D Environments
CRISAM - Inria Sophia Antipolis - Méditerranée
3 GEOMETRICA - Geometric computing
CRISAM - Inria Sophia Antipolis - Méditerranée
Abstract : We address the pixelwise classification of high-resolution aerial imagery. While convolutional neural networks (CNNs) are gaining increasing attention in image analysis, it is still challenging to adapt them to produce fine-grained classification maps. This is due to a well-known trade-off between recognition and localization: the impressive capability of CNNs to recognize meaningful objects comes at the price of losing spatial precision. We here propose an architecture that addresses this issue. It learns features at different levels of detail and also learns a function to combine them. By integrating local and global information in an efficient and flexible manner, it outperforms previous techniques.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [10 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01660754
Contributor : Yuliya Tarabalka <>
Submitted on : Monday, December 11, 2017 - 1:08:51 PM
Last modification on : Thursday, January 11, 2018 - 5:01:43 PM

File

highres.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01660754, version 1

Collections

Citation

Emmanuel Maggiori, Yuliya Tarabalka, Guillaume Charpiat, Pierre Alliez. High-resolution image classification with convolutional networks. IEEE International Geoscience and Remote Sensing Symposium - IGARSS 2017, Jul 2017, Fort Worth, United States. ⟨hal-01660754⟩

Share

Metrics

Record views

252

Files downloads

3296