1 image classification
play

1 Image Classification BVM 2018 Tutorial: Advanced Deep Learning - PowerPoint PPT Presentation

Sep 27, 2022 •577 likes •1.08k views

1 Image Classification BVM 2018 Tutorial: Advanced Deep Learning Methods Jakob Wasserthal, Division of Medical Image Computing Author Division Classification of skin cancer 02.11.16 | vs Esteva et al., Dermatologist-level classification of

  1. 1 Image Classification BVM 2018 Tutorial: Advanced Deep Learning Methods Jakob Wasserthal, Division of Medical Image Computing

  2. Author Division Classification of skin cancer 02.11.16 | vs Esteva et al., Dermatologist-level classification of skin cancer with deep neural networks, Nature, 2017 | Jakob Wasserthal 2

  3. Author Division Classification of skin cancer 02.11.16 | vs benign malignant Esteva et al., Dermatologist-level classification of skin cancer with deep neural networks, Nature, 2017 | Jakob Wasserthal 3

  4. Author Division Classification 02.11.16 | p(benign) 0.98 p(malignant) 0.02 | Jakob Wasserthal 4

  5. Author Division ILSVRC challenge / ImageNet 02.11.16 | top-5 
 second best 
 error 26.2% AlexNet 
 15.3% ZFNet 
 11.2% VGG 
 GoogLeNet 
 7.3% 6.67% Human 
 5.1% ResNet 
 Inception v3 
 DenseNet 
 3.57% 3.5% ~3.5% 2012 2013 2014 2015 2017 | Jakob Wasserthal 5

  6. Author Division ILSVRC challenge / ImageNet 02.11.16 | top-5 
 second best 
 error 26.2% AlexNet 
 15.3% ZFNet 
 11.2% VGG 
 GoogLeNet 
 7.3% 6.67% Human 
 5.1% ResNet 
 Inception v3 
 DenseNet 
 3.57% 3.5% ~3.5% 2012 2013 2014 2015 2017 | Jakob Wasserthal 6

  7. Author Division VGG 02.11.16 | - simple structure - 160M parameters Simonyan et al.,Very deep convolutional networks for 
 large-scale image recognition, arXiv, 2014 He et al., Deep Residual Learning for Image Recognition, arXiv, 2015 | Jakob Wasserthal 7

  8. Author Division ILSVRC challenge / ImageNet 02.11.16 | top-5 
 second best 
 error 26.2% AlexNet 
 15.3% ZFNet 
 11.2% VGG 
 GoogLeNet 
 7.3% 6.67% Human 
 5.1% ResNet 
 Inception v3 
 DenseNet 
 3.57% 3.5% ~3.5% 2012 2013 2014 2015 2017 | Jakob Wasserthal 8

  9. Author Division GoogLeNet 02.11.16 | Inception module Szegedy et al., Going Deeper with Convolutions, arXiv, 2014 | Jakob Wasserthal 9

  10. Author Division GoogLeNet 02.11.16 | stride=1 Szegedy et al., Going Deeper with Convolutions, arXiv, 2014 | Jakob Wasserthal 10

  11. [Width x Height x Nr of Filters] Author Division GoogLeNet 02.11.16 | WxHx(256+256+256+256) = WxHx1024 WxHx256 WxHx256 WxHx256 WxHx256 stride=1 WxHx256 Szegedy et al., Going Deeper with Convolutions, arXiv, 2014 | Jakob Wasserthal 11

  12. [Width x Height x Nr of Filters] Author Division GoogLeNet 02.11.16 | WxHx(256+256+256+256) = WxHx1024 WxHx256 WxHx256 WxHx256 WxHx256 stride=1 WxHx256 WxHx(128+192+96+64) = WxHx480 WxHx192 WxHx96 WxHx64 WxHx128 WxHx128 WxHx32 WxHx256 stride=1 WxHx256 Szegedy et al., 2014 | Jakob Wasserthal 12

  13. Author Division GoogLeNet 02.11.16 | VGG GoogLeNet Data #Parameters dimensions Data #Parameters dimensions 1000*1024=1M 1000 14x14x512 7x 7x512=25.088 1x1x1024 4094 25088*4094=102M 7x7x1024 Szegedy et al., Going Deeper with Convolutions, arXiv, 2014 | Jakob Wasserthal 13

  14. Author Division GoogLeNet 02.11.16 | Inception module - 4M parameters (VGG: 160M) - 22 trained layers Szegedy et al., Going Deeper with Convolutions, arXiv, 2014 | Jakob Wasserthal 14

  15. Author Division Inception v3 - Improvement 1 02.11.16 | Parameters: 5x5-convolution: 5*5=25 2* 3x3-convolution: 2* (3*3)=18 => ~30% less parameters and 
 computations Szegedy et al., Rethinking the Inception Architecture for Computer Vision, arXiv, 2015 | Jakob Wasserthal 15

  16. Author Division Inception v3 - Improvement 1 02.11.16 | Szegedy et al., Rethinking the Inception Architecture for Computer Vision, arXiv, 2015 | Jakob Wasserthal 16

  17. Author Division Inception v3 - Improvement 2 02.11.16 | Parameters: 3x3-convolution: 3*3=9 2* 1x3-convolution: 2* (1*3)=6 => ~33% less parameters and 
 computations Szegedy et al., Rethinking the Inception Architecture for Computer Vision, arXiv, 2015 | Jakob Wasserthal 17

  18. Author Division Inception v3 - Improvement 2 02.11.16 | Szegedy et al., Rethinking the Inception Architecture for Computer Vision, arXiv, 2015 | Jakob Wasserthal 18

  19. Author Division Inception v3 - Improvement 3 02.11.16 | Representational 3x more computations bottleneck Szegedy et al., Rethinking the Inception Architecture for Computer Vision, arXiv, 2015 | Jakob Wasserthal 19

  20. Author Division Inception v3 - Improvement 3 02.11.16 | Representational - No bottleneck 3x more computations bottleneck - 1x computations Szegedy et al., Rethinking the Inception Architecture for Computer Vision, arXiv, 2015 | Jakob Wasserthal 20

  21. Author Division Inception v3 - Improvement 3 02.11.16 | Optimised Inception module Szegedy et al., Rethinking the Inception Architecture for Computer Vision, arXiv, 2015 | Jakob Wasserthal 21

  22. Author Division Inception v3 02.11.16 | - 3.5% top-5 error - 42 Layers - 2.5x number of parameters of GoogLeNet Szegedy et al., Rethinking the Inception Architecture for Computer Vision, arXiv, 2015 | Jakob Wasserthal 22

  23. Author Division ILSVRC challenge / ImageNet 02.11.16 | top-5 
 second best 
 error 26.2% AlexNet 
 15.3% ZFNet 
 11.2% VGG 
 GoogLeNet 
 7.3% 6.67% Human 
 5.1% ResNet 
 Inception v3 
 DenseNet 
 3.57% 3.5% ~3.5% 2012 2013 2014 2015 2017 | Jakob Wasserthal 23

  24. Author Division Classification of skin cancer 02.11.16 | - Inception v3 pretained on ImageNet - Dermatologist-level accuracy Esteva et al., Dermatologist-level classification of skin cancer with deep neural networks, Nature, 2017 | Jakob Wasserthal 24

  25. Author Division Classification of diabetic retinopathy 02.11.16 | - Inception v3 pretained on ImageNet - Expert-level accuracy Gulshan et al., Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs, JAMA, 2016 | Jakob Wasserthal 25

  26. Author Division ILSVRC challenge / ImageNet 02.11.16 | top-5 
 second best 
 error 26.2% AlexNet 
 15.3% ZFNet 
 11.2% VGG 
 GoogLeNet 
 7.3% 6.67% Human 
 5.1% ResNet 
 Inception v3 
 DenseNet 
 3.57% 3.5% ~3.5% 2012 2013 2014 2015 2017 | Jakob Wasserthal 26

  27. Author Division ResNet 02.11.16 | He et al., Deep Residual Learning for Image Recognition, arXiv, 2015 | Jakob Wasserthal 27

  28. Author Division ResNet 02.11.16 | He et al., Deep Residual Learning for Image Recognition, arXiv, 2015 | Jakob Wasserthal 28

  29. Author Division ResNet 02.11.16 | - 152 Layers | Jakob Wasserthal 29

  30. Author Division ResNet 02.11.16 | He et al., Deep Residual Learning for Image Recognition, arXiv, 2015 | Jakob Wasserthal 30

  31. Author Division ILSVRC challenge / ImageNet 02.11.16 | top-5 
 second best 
 error 26.2% AlexNet 
 15.3% ZFNet 
 11.2% VGG 
 GoogLeNet 
 7.3% 6.67% Human 
 5.1% ResNet 
 Inception v3 
 DenseNet 
 3.57% 3.5% ~3.5% 2012 2013 2014 2015 2017 | Jakob Wasserthal 31

  32. Author Division DenseNet 02.11.16 | Huang et al., Densely Connected Convolutional Networks, CVPR, 2017 | Jakob Wasserthal 32

  33. Author Division DenseNet 02.11.16 | Huang et al., Densely Connected Convolutional Networks, CVPR, 2017 | Jakob Wasserthal 33

  34. Author Division Challenges in medical image classification 02.11.16 | - few training data - no RGB images - small lesions - big images - interpretability Source: The Radiology Assistant : Bi-RADS for Mammography and Ultrasound 2013 | Jakob Wasserthal 34

  35. Author Division Interpretability of predictions 02.11.16 | ? p(diabetic) 0.98 p(normal) 0.02 A deep neural network is often considered as a “black box”. | Jakob Wasserthal 35

  36. Author Division Interpretability of predictions 02.11.16 | “What parts of the input image affect the decision?” Gulshan et al., Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs, 2016 | Jakob Wasserthal 36

  37. Author Division Recap: Training via Backpropagation 02.11.16 | w ij x p dog (x) 0.98 p cat (x) 0.02 dc c = − log ( p dog ( x )) dw ij Slides by courtesy of Paul Jäger | Jakob Wasserthal 37

  38. Author Division Saliency maps 02.11.16 | “What parts of the input image affect the decision?” x ij p dog (x) 0.98 p cat (x) 0.02 dp dog ( x ) “backprop into image”: dx ij Slides by courtesy of Paul Jäger | Jakob Wasserthal 38

  39. Author Division Saliency maps 02.11.16 | dp dog ( x ) x ij dx ij Slides by courtesy of Paul Jäger | Jakob Wasserthal 39

  40. Author Division Interpretability of predictions 02.11.16 | Jamaludin et al., SpineNet: Automated classification and evidence visualization in spinal MRIs, Medical image analysis, 2017 | Jakob Wasserthal 40

  41. Author Division 02.11.16 | Questions | Jakob Wasserthal 41

  42. Author Division 02.11.16 | Backup | Jakob Wasserthal 42

  43. Author Division Advanced: Saliency via Perturbation 02.11.16 | “Interpretable Explanations of Black Boxes by Meaningful Perturbation” Ruth et al., arXiv, 2018 Trick: Backprop into a mask d [ w ∗ ( x ∗ m )] m multiplied with the image to = w ∗ x be the “minimal destroying dm region”. | Jakob Wasserthal 43

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Classification Image Classification Set of predefined categories [eg: table, apple, dog, giraffe]

Classification Image Classification Set of predefined categories [eg: table, apple, dog, giraffe]

Day 1 Lecture 2 Classification Image Classification Set of predefined categories [eg: table, apple, dog, giraffe] Binary classification [1, 0] DOG 2 Image Classification 3 Image Classification pipeline Dog 4 Slide credit: Jose M lvarez

339 views • 23 slides
From image classification to object detection Image classification Object detection Image source

From image classification to object detection Image classification Object detection Image source

From image classification to object detection Image classification Object detection Image source Slides from L. Lazebnik What are the challenges of object detection? Images may contain more than one class, multiple instances from the same

1.11k views • 57 slides
Image Analysis System Example: Image Classification System pre feature feature segmentation

Image Analysis System Example: Image Classification System pre feature feature segmentation

Image Analysis System Example: Image Classification System pre feature feature segmentation processing description selection post feature classification classification extraction 1 IIP Image Segmentation Grey-level thresholding

422 views • 15 slides
for Image Classification Qilong Wang ( ) Dalian University of Technology

for Image Classification Qilong Wang ( ) Dalian University of Technology

Codebook-free Single Gaussian for Image Classification Qilong Wang ( ) Dalian University of Technology http://ice.dlut.edu.cn/PeihuaLi/ Image Model for Classification Scene Object Image Model Fine-grained Texture Face

539 views • 43 slides
Object Detection JunYoung Gwak 1 Motivation Image classification Input: Image

Object Detection JunYoung Gwak 1 Motivation Image classification Input: Image

Object Detection JunYoung Gwak 1 Motivation Image classification Input: Image Output: object class 2 Motivation Limitation of classification Multiple classes Location i.e. Object classification assumes Single

1.07k views • 50 slides
Tuning the Performance of Convolutional Neural Network for Image Classification on GPU Agenda

Tuning the Performance of Convolutional Neural Network for Image Classification on GPU Agenda

Tuning the Performance of Convolutional Neural Network for Image Classification on GPU Agenda Adoptions of Image classification or image recognition at Alibaba Easy ways to improve performance of Caffe Further performance

182 views • 15 slides
Natural Language Processing Classification I Dan Klein UC Berkeley 1 2 Classification

Natural Language Processing Classification I Dan Klein UC Berkeley 1 2 Classification

Natural Language Processing Classification I Dan Klein UC Berkeley 1 2 Classification Classification Automatically make a decision about inputs Example: document category Example: image of digit digit Example: image

757 views • 50 slides
for Large-Scale Image Classification Karn Simonyan, Andrea Vedaldi, Andrew Zisserman Visual

for Large-Scale Image Classification Karn Simonyan, Andrea Vedaldi, Andrew Zisserman Visual

Deep Fisher Networks for Large-Scale Image Classification Karn Simonyan, Andrea Vedaldi, Andrew Zisserman Visual Geometry Group, University of Oxford Deep learning achieves excellent performance in image classification. Do hand-crafted image

159 views • 4 slides
Gaze Embeddings for Zero-Shot Image Classification Nour Karessli Zeynep Akata Bernt

Gaze Embeddings for Zero-Shot Image Classification Nour Karessli Zeynep Akata Bernt

Gaze Embeddings for Zero-Shot Image Classification Nour Karessli Zeynep Akata Bernt Schiele Andreas Bulling Presentation by Hsin-Ping Huang and Shubham Sharma Introduction Attributes Standard image classification models fail

758 views • 28 slides
Metric Learning Applied for Automatic Large Image Classification Supervisors TOON CALDERS (PhD) /

Metric Learning Applied for Automatic Large Image Classification Supervisors TOON CALDERS (PhD) /

September, 2014 UPC Metric Learning Applied for Automatic Large Image Classification Supervisors TOON CALDERS (PhD) / ULB SAHILU WENDESON / IT4BI SALIM JOUILI (PhD) /EuraNova Image Database Classification How? Using K-Nearest Neighbor (kNN)

415 views • 40 slides
Hybrid Deep Learning Topology for Image Classification Petru Radu petru.radu@ness.com 27 th

Hybrid Deep Learning Topology for Image Classification Petru Radu petru.radu@ness.com 27 th

15.07.2019 Hybrid Deep Learning Topology for Image Classification Petru Radu petru.radu@ness.com 27 th Summer School on Image Processing, Timisoara 2019 SSIP 2019 Introduction Classical Neural Networks employed in image classification have a

513 views • 33 slides
Statistical NLP Spring 2011 Lecture 11: Classification Dan Klein UC Berkeley Classification

Statistical NLP Spring 2011 Lecture 11: Classification Dan Klein UC Berkeley Classification

Statistical NLP Spring 2011 Lecture 11: Classification Dan Klein UC Berkeley Classification Automatically make a decision about inputs Example: document category Example: image of digit digit Example: image of object

609 views • 48 slides
Image Classification with Fashion-MNIST and CIFAR-10 Khoi Hoang California State University,

Image Classification with Fashion-MNIST and CIFAR-10 Khoi Hoang California State University,

Image Classification with Fashion-MNIST and CIFAR-10 Khoi Hoang California State University, Sacramento Problem Statement Image Classification is one of the most fundamental problem in the field of machine learning. Has many area of

639 views • 10 slides
Bag-of-features models for category classification for category classification Cordelia Schmid

Bag-of-features models for category classification for category classification Cordelia Schmid

Bag-of-features models for category classification for category classification Cordelia Schmid Category recognition Category recognition Image classification: assigning a class label to the image Image classification: assigning a class

998 views • 75 slides
Lecture 7: Multiclass Classification Princeton University COS 495 Instructor: Yingyu Liang

Lecture 7: Multiclass Classification Princeton University COS 495 Instructor: Yingyu Liang

Machine Learning Basics Lecture 7: Multiclass Classification Princeton University COS 495 Instructor: Yingyu Liang Example: image classification indoor Indoor outdoor Example: image classification (multiclass) ImageNet figure borrowed from

807 views • 28 slides
AMMI Introduction to Deep Learning 7.2. Networks for image classification Fran cois

AMMI Introduction to Deep Learning 7.2. Networks for image classification Fran cois

AMMI Introduction to Deep Learning 7.2. Networks for image classification Fran cois Fleuret https://fleuret.org/ammi-2018/ Thu Sep 6 15:26:25 CAT 2018 COLE POLYTECHNIQUE FDRALE DE LAUSANNE Image classification, standard convnets

827 views • 71 slides
WEL ELCOME OME All Program Directors & FM Site Directors Meeting Fr Frid iday ay, , Ma

WEL ELCOME OME All Program Directors & FM Site Directors Meeting Fr Frid iday ay, , Ma

WEL ELCOME OME All Program Directors & FM Site Directors Meeting Fr Frid iday ay, , Ma May y 25 25, 20 2018 18 AGENDA Welcome Awards & Thanks PAAC Update Charles Mickle Fellowship Address Accreditation

1.19k views • 117 slides
Introduction CSCE CSCE Sometimes a single classifier (e.g., neural network, 478/878 478/878

Introduction CSCE CSCE Sometimes a single classifier (e.g., neural network, 478/878 478/878

Introduction CSCE CSCE Sometimes a single classifier (e.g., neural network, 478/878 478/878 Lecture 7: Lecture 7: decision tree) wont perform well, but a weighted CSCE 478/878 Lecture 7: Bagging and Bagging and Boosting Boosting

480 views • 4 slides
Data Mining with Weka Class 4 Lesson 1 Classification boundaries Ian H. Witten Department of

Data Mining with Weka Class 4 Lesson 1 Classification boundaries Ian H. Witten Department of

Data Mining with Weka Class 4 Lesson 1 Classification boundaries Ian H. Witten Department of Computer Science University of Waikato New Zealand weka.waikato.ac.nz Lesson 4.1 Classification boundaries Class 1 Getting started with Weka Lesson 4.1

679 views • 45 slides
NANASP Welcome NANASP is very pleased to have this follow-up webinar to our conference session

NANASP Welcome NANASP is very pleased to have this follow-up webinar to our conference session

8/22/2019 Leveraging New Opportunities to Address the Social Determinants of Health NANASP Webinar August 20, 2019 8/22/2019 1 NANASP Welcome NANASP is very pleased to have this follow-up webinar to our conference session on social

456 views • 19 slides
Big Data in Pharmaceutical Industry (Novartis) Jean-Michel Gaullier, PhD Sollentuna, August 3 rd

Big Data in Pharmaceutical Industry (Novartis) Jean-Michel Gaullier, PhD Sollentuna, August 3 rd

Big Data in Pharmaceutical Industry (Novartis) Jean-Michel Gaullier, PhD Sollentuna, August 3 rd 2015 Future Cloud Summer School Pharmaceutical Industry 2 | Presentation Title | Presenter Name | Date | Subject | Business Use Only Future Cloud

672 views • 30 slides
Data-Driven Research Program Data-Driven Research Program Linked Longitudinal Retrospective

Data-Driven Research Program Data-Driven Research Program Linked Longitudinal Retrospective

VICTORIAN COMPREHENSIVE CANCER CENTRE Data-Driven Research Program Data-Driven Research Program Linked Longitudinal Retrospective Health care continuum Secondary use Bottom-up Data-Driven Research Program Hospital Primary Care Population

380 views • 22 slides
Administration Reading assignment Computing and the Supplementary Readings Developing

Administration Reading assignment Computing and the Supplementary Readings Developing

Administration Reading assignment Computing and the Supplementary Readings Developing World CSEP 590B, Spring 2008 Authentication Lecture 3 - Telemedicine User: csep590b Password: student Richard Anderson Todays

254 views • 7 slides
Information Systems for Administration, 1 09/2009

Information Systems for Administration, 1 09/2009

Information Systems for Administration, 1 09/2009 Clinical Documentation and Quality Assurance in an Austrian Disease Management Programme MIE 2009, Sarajevo Peter Beck JOANNEUM RESEARCH, Graz,

345 views • 20 slides

More recommend