Deep Learning For Aerospace Applications - Teratec

1y ago
28 Views
2 Downloads
4.44 MB
35 Pages
Last View : 13d ago
Last Download : 4m ago
Upload by : Jamie Paz
Transcription

1 Deep learning for aerospace applications Alexandre Boulch 2017

2 Deep Learning Lee Sedol 2015/10 Ke Jie 2017/05 2017

Deep Learning 3 Personal assistant Personalised learning Recommendations Réponse automatique Deep learning and Big data for cardiology 2017

Deep Learning 2017 4

5 Overview Machine Learning Deep Learning DeLTA 2017

6 AI The science and engineering of making intelligent machines. Logical, search, pattern recognition, planning, inference, learning from experience. 2017

7 AI Intelligent machines Machine Learning Learning from experience Training data Input 2017 Tuning parameters Model with parameters Output

Machine learning starts in the 60’s 2017 8

9 Overview Machine Learning Deep Learning DeLTA 2017

Deep learning 2017 10

Deep learning 2017 11

Deep learning Datasets Competition Emulation New optimizers New neural layers New architectures . Better Faster Stronger 2017 12

13 Deep learning 2017 2017

Deep learning AI Intelligent machines Machine Learning Learning from experience Deep learning Auto-learning Deep neural networks 2017 14

15 Deep learning Feature extraction Decision Expert Data Machine learning Knowledge about statistics Knowledge about data and application 2017

Deep learning: a massively data driven approach Features and decision Data Deep neural networks Network suited for applications Data knowledge 2017 Statistics, optimization 16

Deep learning: a massively data driven approach Features and decision Data Deep neural networks Network suited for applications Data knowledge 2017 Statistics, optimization 17

18 Machine learning at ONERA DAAA DMPE DEMR DPhiEE 2017 DOTA DMAS DTIS

19 Deep learning at ONERA DAAA DMPE DEMR DPhiEE DOTA DMAS DTIS A. Chan-Hon-Tong, S. Herbin, B. Le Saux, A. Boulch . 2017

20 3 x ( 3x3 conv. ReLU ) RGB Composite Semantic Map labeling Aerial images, multimodal (RGB, IR, DSM, .) Fusion networks PhD Nicolas Audebert (nicolas.audebert.at) 2017

21 Point cloud labeling 2 x ( 3x3 conv. ReLU ) Max Pooling 2x2 2 x ( 3x3 conv. ReLU ) Max Pooling 2x2 3 x ( 3x3 conv. ReLU ) Max Pooling 2x2 3 x ( 3x3 conv. ReLU ) Max Pooling 2x2 3 x ( 3x3 conv. ReLU ) Deconv. 3x3 Concatenation VGG 16 Batch Norm 3 x ( 3x3 conv. ReLU ) Deconv. 3x3 Batch Norm 3x(3x3 conv. ReLU ) Deconv. 3x3 Batch Norm 2 x ( 3x3 conv. ReLU ) Deconv. 3x3 Batch Norm 2 x ( 3x3 conv. ReLU ) Leader on Semantic 8 LIDAR dataset Transfer to photogrammetry Code available online (DeLTA website) 2017

Detection RGB and Depth for person detection improvment PhD Joris Guery jorisguerry.fr 2017 22

Detection Detection in low resolution Images Exploitation of images Sequences for detection Juliette Chataigner (Intern) 2017 23

Depth from defocus Sensor specific processing Depth from de focus. PhD Macella Carvalho 2017 24

25 Zero Shot Learning Zero-Shot Learning via Visual Abstraction Stanislaw Antol, Larry Zitnick, Devi Parikh Zero Shot Learning Learning based on attributes PhD Maxime Bucher 2017

26 Overview Machine Learning Deep Learning DeLTA 2017

27 Deep learning for aerospace ONERA 2017

28 Deep learning for aerospace ONERA Frameworks Data R&D for aerospace and defense Development 2017

29 LE PRF DeLTA Core skills Deep Lab Applications 2017

30 Databases Domain adaptation State of the art Core skills New architectures 2017

31 LE PRF DeLTA Academics and industrials Tutorials Improve ONERA research exposure Networks Code Datasets Software and help to solve problems with machine learning Deep Lab 2017 Results

32 Database Learning Database code Problem code Networks Validation How To Deep Lab 2017 Generic code Network base Practical case

33 Core skills Deep Lab DOTA Atmospheric Corr. Applications DAAA Fluid mecanics DEMR Detection and recognition DTIS Robotics 2017 DMAS Material mecanics

34 4 year project Capitalize Share Reproduce Core skills Move forward Innovate delta-onera.github.io 2017 Deep Lab Applications Transfer Experiment Evaluate

35 “We chose it because we deal with huge amounts of data. Besides, it sounds really cool.” Larry Page - Google 2017

Deep Learning Personal assistant Personalised learning Recommendations Réponse automatique Deep learning and Big data for cardiology. 4 2017 Deep Learning. 5 2017 Overview Machine Learning Deep Learning DeLTA. 6 2017 AI The science and engineering of making intelligent machines.

Related Documents:

Bruksanvisning för bilstereo . Bruksanvisning for bilstereo . Instrukcja obsługi samochodowego odtwarzacza stereo . Operating Instructions for Car Stereo . 610-104 . SV . Bruksanvisning i original

10 tips och tricks för att lyckas med ert sap-projekt 20 SAPSANYTT 2/2015 De flesta projektledare känner säkert till Cobb’s paradox. Martin Cobb verkade som CIO för sekretariatet för Treasury Board of Canada 1995 då han ställde frågan

service i Norge och Finland drivs inom ramen för ett enskilt företag (NRK. 1 och Yleisradio), fin ns det i Sverige tre: Ett för tv (Sveriges Television , SVT ), ett för radio (Sveriges Radio , SR ) och ett för utbildnings program (Sveriges Utbildningsradio, UR, vilket till följd av sin begränsade storlek inte återfinns bland de 25 största

Hotell För hotell anges de tre klasserna A/B, C och D. Det betyder att den "normala" standarden C är acceptabel men att motiven för en högre standard är starka. Ljudklass C motsvarar de tidigare normkraven för hotell, ljudklass A/B motsvarar kraven för moderna hotell med hög standard och ljudklass D kan användas vid

LÄS NOGGRANT FÖLJANDE VILLKOR FÖR APPLE DEVELOPER PROGRAM LICENCE . Apple Developer Program License Agreement Syfte Du vill använda Apple-mjukvara (enligt definitionen nedan) för att utveckla en eller flera Applikationer (enligt definitionen nedan) för Apple-märkta produkter. . Applikationer som utvecklas för iOS-produkter, Apple .

Grade Level/4-H Experince Pre-Flight (1) Aerospace Lift-Off (2) Aerospace Reaching New Heigts (3) Aerospace Pilot in Command (4) Aerospace Flight Crew Aerospace Leaders Launching Youth Aerospace Programs Aerospace Leaders Gifts of Gold: Seeds, Stalks, & Science Agriculture Gifts of Gold: Food

Deep Learning: Top 7 Ways to Get Started with MATLAB Deep Learning with MATLAB: Quick-Start Videos Start Deep Learning Faster Using Transfer Learning Transfer Learning Using AlexNet Introduction to Convolutional Neural Networks Create a Simple Deep Learning Network for Classification Deep Learning for Computer Vision with MATLAB

quality results of AGMA Class 10 with a double cut cycle and AGMA Class 9 with a single cut cycle can be expect-ed. The 100H weighs 7,100 lbs with a cube size of 78” X 64” X 72” high. For more information, contact Bourn & Koch at (815) 965-4013, Boeing F/A-18E/F Super Hornet Completes First AESA Flight In a press release dated August 13, 2003, it was announced that the Boeing F/A-18E/F .