no code implementations • 11 Jul 2023 • Johann Lussange, Mulin Yu, Yuliya Tarabalka, Florent Lafarge
We here propose a method for urban 3D reconstruction named KIBS(\textit{Keypoints Inference By Segmentation}), which comprises two novel features: i) a full deep learning approach for the 3D detection of the roof sections, and ii) only one single (non-orthogonal) satellite raster image as model input.
1 code implementation • CVPR 2021 • Nicolas Girard, Dmitriy Smirnov, Justin Solomon, Yuliya Tarabalka
While state of the art image segmentation models typically output segmentations in raster format, applications in geographic information systems often require vector polygons.
1 code implementation • NeurIPS 2019 • Guillaume Charpiat, Nicolas Girard, Loris Felardos, Yuliya Tarabalka
We first exhibit a multimodal image registration task, for which a neural network trained on a dataset with noisy labels reaches almost perfect accuracy, far beyond noise variance.
no code implementations • 17 Aug 2020 • Gael Kamdem De Teyou, Yuliya Tarabalka, Isabelle Manighetti, Rafael Almar, Sebastien Tripod
Experimental results show that the temporal information provided by time series images allows increasing the accuracy of land cover classification, thus producing up-to-date maps that can help in identifying changes on earth.
no code implementations • 13 May 2020 • Onur Tasar, Alain Giros, Yuliya Tarabalka, Pierre Alliez, Sébastien Clerc
We propose a novel approach, coined DAugNet, for unsupervised, multi-source, multi-target, and life-long domain adaptation of satellite images.
2 code implementations • 30 Apr 2020 • Nicolas Girard, Dmitriy Smirnov, Justin Solomon, Yuliya Tarabalka
While state of the art image segmentation models typically output segmentations in raster format, applications in geographic information systems often require vector polygons.
no code implementations • 14 Apr 2020 • Onur Tasar, Yuliya Tarabalka, Alain Giros, Pierre Alliez, Sébastien Clerc
However, these methods have limited practical real world applications, since usually one has multiple source domains with different data distributions.
no code implementations • 14 Feb 2020 • Onur Tasar, S. L. Happy, Yuliya Tarabalka, Pierre Alliez
Although convolutional neural networks have been proven to be an effective tool to generate high quality maps from remote sensing images, their performance significantly deteriorates when there exists a large domain shift between training and test data.
1 code implementation • 30 Jul 2019 • Onur Tasar, S. L. Happy, Yuliya Tarabalka, Pierre Alliez
Due to the various reasons such as atmospheric effects and differences in acquisition, it is often the case that there exists a large difference between spectral bands of satellite images collected from different geographic locations.
1 code implementation • 12 Mar 2019 • Nicolas Girard, Guillaume Charpiat, Yuliya Tarabalka
In machine learning the best performance on a certain task is achieved by fully supervised methods when perfect ground truth labels are available.
no code implementations • 29 Oct 2018 • Onur Tasar, Yuliya Tarabalka, Pierre Alliez
The key points of the proposed approach are adapting the network to learn new as well as old classes on the new training data, and allowing it to remember the previously learned information for the old classes.
no code implementations • ECCV 2018 • Armand Zampieri, Guillaume Charpiat, Nicolas Girard, Yuliya Tarabalka
We tackle here the problem of multimodal image non-rigid registration, which is of prime importance in remote sensing and medical imaging.
no code implementations • 27 Feb 2018 • Armand Zampieri, Guillaume Charpiat, Yuliya Tarabalka
We tackle here the problem of multimodal image non-rigid registration, which is of prime importance in remote sensing and medical imaging.
no code implementations • 8 Dec 2016 • Seong-Gyun Jeong, Yuliya Tarabalka, Nicolas Nisse, Josiane Zerubia
We propose a novel tree-like curvilinear structure reconstruction algorithm based on supervised learning and graph theory.
no code implementations • 7 Nov 2016 • Emmanuel Maggiori, Yuliya Tarabalka, Guillaume Charpiat, Pierre Alliez
We establish the desired properties of an ideal semantic labeling CNN, and assess how those methods stand with regard to these properties.
no code implementations • 11 Aug 2016 • Emmanuel Maggiori, Guillaume Charpiat, Yuliya Tarabalka, Pierre Alliez
Instead, our goal is to directly learn the iterative process itself.