I am a computer vision, machine learning and remote sensing researcher. I hold a PhD in Computer Science prepared at ONERA and IRISA, supervised by Sébastien Lefèvre and Bertrand Le Saux. My thesis was about teaching computers how to look at airborne images to create beautiful maps. My research now focuses on deep neural networks (deep learning) for artificial perception, ranging from document image analysis to remote sensing and Earth Observation. Specifically, I am currently working on multimodal learning for image/text understanding in administrative documents.
I am interested in machine learning, algorithms and Python.
July 2019: I will be at APIA 2019 in Toulouse from July 1st to July 5th to present our work on multi-modal text/image classification with deep nets. Feel free to come for a chat!
May 2019: Our paper on multi-modal text/image deep networks for document image classification has been accepted to APIA 2019 in Toulouse.
April 2019: I will be presenting at the GdR ISIS meeting on weakly and semi-supervised learning for image and video classification. My talk will detail some of the work I did at Quicksign on image/text clustering for document recognition.
April 2019: Our review on deep convolutional and recurrent neural networks for hyperspectral image classification has ben accepted for the IEEE Geoscience and Remote Sensing special issue on hyperspectral data. Preprint here.
October 2018: I successfully defended my PhD thesis! The manuscript (in french) is available here with slides.
July 2018: I was at IGARSS'18 in Valencia, where I presented our work on generative adversarial network for hyperspectral samples synthesis. You can find the code here!
March 2018: We have one paper accepted for IGARSS 2018 on generative adversarial networks for hyperspectral data synthesis. We'll also appear on the Inria Aerial Image Labeling benchmark write-up on building extraction.
January 2018: I ported the code of our deep network for aerial/satellite semantic segmentation to PyTorch for an easier use: fork it on GitHub!
November 2017: Our latest journal paper on data fusion for remote sensing data using deep fully convolutional networks is out !
July 2017: I was at CVPR 2017 for the Earthvision workshop, where I presented our work on semantic mapping using deep nets and OpenStreetMap data.
June 2017: I collaborated with the LISTIC team on using deep nets to perform semantic segmentation on Sentinel-2 images. This work will be presented at IGARSS'17 in Forth Worth, Texas.
June 2017: I presented at ORASIS 2017 our work on data fusion with deep networks for remote sensing (slides).
May 2017: Our submission on joint deep learning using optical and OSM data for semantic mapping of aerial/satellite images has been accepted to the EarthVision 2017 CVPR Workshop !
April 2017: Our Remote Sensing journal paper on vehicle segmentation for detection and classification is out in open access on the MPDI website.
March 2017: My colleague Alexandre Boulch will present the SnapNet architecture for semantic segmentation of unstructured point clouds at Eurographics 3DOR workshop. It is the current state-of-the-art on the Semantic3D dataset (code).
March 2017: Our paper on data fusion for remote sensing using deep nets won the 2nd best student paper award at JURSE 2017 ! Slides and poster are available.
Februrary 2017: The code of the deep network we used for the ISPRS Vaihingen 2D Semantic Labeling Challenge is out on Github !
January 2017: We will present two invited papers at JURSE 2017 !
November 2016: I will be at ACCV'16 in Taipei to present our poster on semantic segmentation of Earth Observation using multi-scale and multimodal deep networks.
October 2016: I will be at PyCon-fr (the French Python conference) to speak about deep learning using Python (slides (in French) and video (in French, too)).
September 2016: Our paper on the use of deep networks for object-based image analysis of vehicles in the ISPRS dataset has been distinguished by the "Best Benchmarking Contribution Award" at GEOBIA 2016 !
September 2016: I will be at GEOBIA 2016 in Enschede to talk about our work on object-based analysis of cars in remote sensing images using deep learning.
September 2016: Our paper on semantic segmentation for Earth Observation was accepted at ACCV'16 for a poster presentation. Check out the state-of-the-art results on the ISPRS Vaihingen 2D Semantic Labeling Challenge !
July 2016: I will be at IGARSS'16 in Beijing to present our work on superpixel-based semantic segmentation of aerial images.
April 2016: Our paper on region-based classification of remote sensing images using deep features has been accepted at IGARSS'16 for an oral presentation.
October 2015: I started as a PhD student at ONERA and IRISA.
PublicationsMost of my papers should be in open access on HAL or on arXiv.
- "Deep learning for classification of hyperspectral data: a comparative review", Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre, IEEE Geosciences and Remote Sensing Magazine, 2019 (to appear).
- "Beyond RGB: Very High Resolution Urban Remote Sensing With Multimodal Deep Networks", Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre, ISPRS Journal of Photogrammetry and Remote Sensing, Elsevier, 2018.
- "Segment-before-Detect: Vehicle Detection and Classification through Semantic Segmentation of Aerial Images", Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre, Remote Sensing, MDPI, 2017.
- "Generative Adversarial Networks for Realistic Synthesis of Hyperspectral Samples", Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre, IGARSS, Valencia, 2018.
- "Couplage de données géographiques participatives et d'images aériennes par apprentissage profond", Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre, GRETSI, Juan-les-Pins, 2017.
- "Joint Learning from Earth Observation and OpenStreetMap Data to Get Faster Better Semantic Maps", Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre, EarthVision - CVPR Workshop, Honolulu, 2017 (poster).
- "Unstructured point cloud semantic labeling using deep segmentation networks", Alexandre Boulch, Bertrand Le Saux, Nicolas Audebert, Eurographics 3DOR, Lyon, 2017.
- "Fusion of Heterogeneous Data in Convolutional Networks for Urban Semantic Labeling", Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre, JURSE, Dubai, 2017 (slides, poster).
- "Semantic Segmentation of Earth Observation Data Using Multimodal and Multi-scale Deep Networks", Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre, ACCV, Taipei, 2016 (poster).
- "On the usability of deep networks for object-based image analysis", Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre, GEOBIA, Enschede, 2016 (slides).
- "How useful is region-based classification of remote sensing images in a deep learning framework ?", Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre, IGARSS, Beijing, 2016 (slides).
- "Structural classifiers for contextual semantic labeling of aerial images". Hicham Randrianarivo, Bertrand Le Saux, Nicolas Audebert, Michel Crucianu, Marin Ferecatu, Big Data from Space (BiDS), Tenerife, 2016.
- "Multimodal deep networks for text and image-based document classification", Nicolas Audebert, Catherine Herold, Kuider Slimani, Cédric Vidal, APIA (Applications Pratiques de l'Intelligence Artificielle), Toulouse, 2019 (slides).
- "Réseaux de neurones profonds et fusion de données pour la segmentation sémantique d'images aériennes", Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre, ORASIS 2017 (Journées des jeunes chercheurs en vision par ordinateur), Colleville-sur-Mer, 2017 (slides).
- "Deep Learning for Remote Sensing". Nicolas Audebert, Alexandre Boulch, Adrien Lagrange, Bertrand Le Saux, Sébastien Lefèvre, 16th ONERA-DLR Aerospace Symposium (ODAS), Oberpfaffenhofen, 2016.
- "Deep learning for aerial cartography" (poster). Nicolas Audebert, Bertrand Le Saux, Sébastien Lefèvre, Statlearn Workshop, Vannes, 2016.
- QS-OCR dataset: this dataset is a collection of 400,000 text documents from the RVL-CDIP and Tobacco3482 image document datasets processed by the Tesseract-OCR engine. It can be used for document classification (NLP setting) or in conjunction with the RVL-CDIP/Tobacco datasets for a multimodal text/image classification task.
- GDR ISIS @ CNAM : Automated document image annotation using text/image clustering (10/05/2019)
- Deep learning workshop @ SAGEO: Deep learning for cartography (06/11/2018)
- GDR ISIS @ CNES : Deep learning for hyperspectral data: a review (18/10/2018)
- PhD defense @ ONERA : Classification of big remote sensing data (17/10/2018)
- IGARSS @ Valencia : GAN for hyperspectral data augmentation (24/07/2018)
- PhD Students Day @ ONERA : Classification of Big Remote Sensing Data (05/02/2018)
- MATHIAS Conference @ TOTAL : Deep 3D CNN for Hyperspectral Image Classification (27/10/2017)
- Data science for geosciences workshop @ EAGE : Deep 3D Convolutional Neural Networks for Classification of Hyperspectral Data (12/06/2017)
- GDR ISIS @ Télécom ParisTech : Deep Learning for Classification of Hyperspectral Data (01/06/2017
- Seminar @ IGN : Deep Learning for Multimodal Remote Sensing and Object Detection (21/03/2017)
- MS SIO @ CentraleSupélec : Deep Learning for Multimodal Remote Sensing and Object Detection (10/03/2017)
- JURSE @ Dubai : Fusion of Heterogeneous Remote Sensing Data Using Deep Convolutional Networks (07/03/2017)
- PhD Students Day @ ONERA : Classification of Big Remote Sensing Data (30/01/2017)
- Seminar on GPU computing @ ONERA : GPU computing for deep learning (12/01/2017)
- PhD Students Day @ IRISA : Classification of Big Remote Sensing Data (08/12/2016)
- MATHIAS Conference @ TOTAL : Object-based Classification of Vehicles in Aerial Images Using Deep Neural Networks (28/10/2016)
- PyCon-FR : Deep learning with Python (15/10/2016)
- GEOBIA @ Enschede : On the Usability of Deep Networks for Object-based Image Analysis (15/09/2016)
- IGARSS @ Beijing : How useful is region-based classification of remote sensing data in a deep learning framework? (13/07/2016)
- ODAS Conference @ ONERA & DLR : Deep learning for remote sensing (22/06/2016)
- Scientific workshop on image processing @ TOTAL : Deep learning for remote sensing (15/06/2016)
ENPC : First year main course of C++ programming
ENPC : First year algorithmic and data structures
|January 2019 - today||Research scientist||Quicksign||Deep learning for document image analysis: image and text classification, OCR, semi-supervised & continuous learning.|
|April - September 2015||Research internship||TUM - Computer Vision Lab||Deep learning for facial expression recognition|
|June - September 2014||Software Engineer (intern)||Withings||Web development of an incentive platform for quantified self (PHP/Backbone.js)|
|2012 - 2015||Member||Supélec Rézo||Sysadmin and software development for the organization providing Internet access to 700 students|
|2015 - 2018||PhD graduate -
"Classification of big remote sensing data"
Deep learning for image processing and Earth Observation
|ONERA, The French Aerospace Lab
Institute for Research in Computer Science and Random Systems (IRISA)
|2014 - 2015||Master Human-Computer Interaction||Université Paris-Sud|
|2012 - 2015||Engineering student - Computer Science||Supélec|
|2012 - 2013||Bachelor Mathematics||Université Paris-Sud|
Skills :Programming :
- Advanced: Python (numpy, scipy, PyTorch, TensorFlow/Keras, sklearn, skimage)
- Beginner: C, Java
- Git, GIMP, Libre Office, LaTeX, GNU/Linux
- French (maternal)
- English (full proficiency)
- Japanese (beginner)
- German (beginner)
I graduated in 2015 from Supélec, one of France's top engineering school, with a focus on Computer Science. I also graduated from the université Paris-Sud with a MSc in Human-Computer Interaction. While I was an engineering student, I was head of the Supélec Rézo organization that provides Internet access to all Supélec' students.
I did my final intership in the Computer Vision lab at TUM under Pr. Daniel Cremers supervision. I worked on deep learning for facial expression recognition.
From October 2015 to October 2018 I was a PhD candidate at ONERA.
Since January 2019 I am a research scientist at Quicksign.
I have an Erdős number of 4.
My Kardashian index is of 0.61.