What is beyond using Tensorflow, GPU or TPU to process images seamlessly? Do we have a silver bullet for image processing? Over the years, image processing has picked up a different level of attraction. Everyone can think about its ease of usability because it has become a reality now. We have started seeing how Residual Neural Network architecture is being used for different cases and not only that, how Residual Neural network is being tweaked to solve different problems. Along with tweaking the ResNet, preprocessing is also being improved to support different architecture for this matter. Everyone has almost become cyborg already with mobile phones in our hands and apparently until human beings bring the AI/ML to the phones completely they are not taking any rest. We are going to see the development of different architecture and algorithms around running AI/ML on low configuration devices. In this session, we are going to talk about different research papers submitted for these matters and some implementations for the same as well.

1. Finding the best solution for
Image Processing
Presented By : Pranjut Gogoi & Shubham
Goyal

2. 2
Our Agenda
01 Image Processing history
02 Different Approaches
03 Residual Neural Networks
04 Performances
05 Ongoing researches

3. 3
About Knoldus MachineX
MachineX is a group of data wizards.
We are a team of Data Scientist and engineers with a
product mindset who deliver competitive business
advantage.

4. 4
An Intelligent
Meeting Assistant
Application
Record Videos
View DashBoard

5. 5

6. 6
An Intelligent
marketing tool
FishEye
FishEye

7. 7
Machine learning library
in scala
KSAI

8. 8
Enable organizations to
capture new value
and business capabilities
Innovation Labs
Consistently blogging, to
share our knowledge,
research
Blogs
Deeplearning, Coursera,
Stanford certified
professionals
Certifications
Insight & perspective to help
you to make right business
decisions
TOK Sessions
It’s great to contribute back
to the community. We
continuously advance open
source technologies to meet
demanding business
requirements.
Open Source
Contribution

9. Finding the best solution for
Image Processing

10. 10
Image processing

11. 11
Image processing History
Traditional way

12. 12
Traditional Way
Traditional pipeline for image classification involves two
modules
● Feature extraction
● Classification

13. 13
Problems
The problem with this pipeline
● Feature extraction cannot be tweaked according to
the classes and images
● Completely different from how we humans learn to
recognize things.

14. Convolutional Neural Network
(CNN, or ConvNet)

15. 15
● Convolutional base
● Classifier

16. Transfer learning

17. 17
The Application of
skills, knowledge,
and/or attitudes that
were learned in one
situation to another
learning situation
transfer learning is usually
expressed through the use of
pre-trained models

18. 18

19. 19
Problems
The problem was
● less learned rate in each generation
● Number of knowledge amount passed down was
less

20. 20

21. 21
Difference

22. Understanding various architectures of
Convolutional Networks
ResNet, AlexNet, VGGNet, Inception

23. 23
ImageNet Large Scale Visual Recognition Challenge
(ILSVRC)
CNN architectures of ILSVRC
top competitors

24. 24
AlexNet
● 5 Convolutional (CONV) layers and 3 Fully Connected (FC) layers
● 62 million trainable variables

25. 25
AlexNet

26. 26
AlexNet
● Data augmentation is carried out to reduce overfitting
● Used Relu which achieved 25% error rate about 6 times faster
than the same network with tanh nonlinearity.
● AlexNet introduced Local Response Normalization (LRN) to
help with the vanishing gradient problem

27. 27
VGGNet
● VGG16 has a total of 138 million parameters
● Conv kernels are of size 3x3 and maxpool kernels are of size 2x2 with
stride of two

28. 28
VGGNet

29. 29
VGGNet
● It is painfully slow to train.
● Spatial pooling is carried out by five max-pooling layers, which
follow some of the conv. layers

30. 30

31. ResNet : Deep Residual learning

32. 32
Hierarchical Features and role of Depth
● Low, Mid , and High-level features
● More layers enrich the “levels” of the features
● Previous ImageNet models have depths of 16 and 30
layers

33. Is learning better networks as easy as
stacking more layers ?

34. 34
Adding layers to deep
Convolutional neural nets

35. 35
Construction Insight
● Consider a shallow architecture and its deeper
counterpart
● The deeper model would would just need to copy the
shallower model with identity mapping
● Construction solution suggests that a deeper model
should produce no higher training error that its shallow
counterpart

36. 36
Residual Functions
● We explicitly reformulate the layers as learning residual functions
with reference to the layer inputs, instead of learning
unreferenced functions
● H[x] = F[x] + x

37. 37

38. 38
Residual vs Plain

39. 39
Experiment
● 152 layer Layers on ImageNet
○ 8* Deeper than VGGNet
○ Less parameters
● ResNet achieve 3.57% error on Imagenet test
○ 1st place in ILSVRC

40. 40
Results
● AlexNet and ResNet-152, both have about 60M parameters but there is
about 10% difference in their top-5 accuracy
● VGGNet not only has a higher number of parameters and FLOP as compared
to ResNet-152, but also has a decreased accuracy
● Training an AlexNet takes about the same time as training Inception (10
times less memory requirements)

41. 41
Clinic Assistant
● Notebook http://bit.ly/2D2LOQT
● Web App https://virtual-clinic.onrender.com

42. 42
History and its importance
● Origin of CNN(1980s-1999)
● Stagnation of CNN(Early 2000)
● Revival of CNN (2006-2011)
● Rise of CNN (2012-2014)
● Rapid increase in Architectural Innovations (2015-present)
● Important because we are not done yet.

43. 43
Taxonomy of deep CNN

44. 44
Spatial Exploitation based CNNs
● LeNet
● AlexNet
● ZefNet
● VGG
● GoogleNet

45. 45
Depth based CNNs
● Highway Networks
● ResNet
● Inception-V3/V4
● Inception-ResNet
● ResNext

46. 46
Multi-path based CNNs
● Highway Nets
● ResNet
● DenseNet

47. 47
Width based CNNs
● WideResNet
● Pyramidal Net
● Xception
● Inception Family

48. 48
Feature map exploitation based CNNs
● Squeeze and Excitation
● Competitive Squeeze and Excitation

49. 49
Channel boosting
● Channel boosted using TL

50. 50
Attention based CNNs
● Residual Attention Neural Network
● Convolutional block attention
● Concurrent Squeeze and Excitation

51. 51
Improvement summary
● Learning capacity of CNN is significantly improved over
the years by exploiting depth and other structural
modifications.
○ Activation, loss function, optimization, regularization,
learning algorithms, and restructuring of processing
units.
● Major improvement on CNN
○ Main boost in CNN performance has been achieved by
replacing the conventional layer structure with blocks

52. 52
Challenge Exists
● Deep NN are generally like a black box and thus may lack
in interpretation and explanation
● Each layer of CNN automatically tries to extract better and
problem specific features related to the task
● Deep CNNs are based on supervised learning
mechanism, and therefore, availability of a large and
annotated data is required for its proper learning
● Hyperparameter selection highly influences the
performance of CNN
● Efficient training of CNN demands powerful hardware
resources such as GPUs.

53. 53
Future of research
● Ensemble learning
● Attention modeling
● Generative learning

54. 54
References
● [1]. A. Krizhevsky, I. Sutskever, and G. E. Hinton. Imagenet classification with deep convolutional
neural networks. In Advances in neural information processing systems,pages 1097–1105,2012.
● [2]. K. He, X. Zhang, S. Ren, and J. Sun. Deep residual learning for image recognition. arXiv preprint
arXiv:1512.03385,2015.
● [3]. K. Simonyan and A. Zisserman. Very deep convolutional networks for large-scale image
recognition. arXiv preprint arXiv:1409.1556,2014.
● [4]. C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A.
Rabinovich. Going deeper with convolutions. In Proceedings of the IEEE Conference on
Computer Vision and Pattern Recognition,pages 1–9,2015.
● https://arxiv.org/pdf/1901.06032.pdf

55. 55
Thank You