https://telecombcn-dl.github.io/2018-dlcv/ Deep learning technologies are at the core of the current revolution in artificial intelligence for multimedia data analysis. The convergence of large-scale annotated datasets and affordable GPU hardware has allowed the training of neural networks for data analysis tasks which were previously addressed with hand-crafted features. Architectures such as convolutional neural networks, recurrent neural networks and Q-nets for reinforcement learning have shaped a brand new scenario in signal processing. This course will cover the basic principles and applications of deep learning to computer vision problems, such as image classification, object detection or image captioning.

1. Xavier Giro-i-Nieto
xavier.giro@upc.edu
Associate Professor
Universitat Politecnica de Catalunya
Technical University of Catalonia
Language and Vision
Day 4 Lecture 4
#DLUPC
http://bit.ly/dlcv2018

2. 2
Acknowledgments
Antonio
Bonafonte
Santiago
Pascual

3. 3
Acknowledgments
Marta R. Costa-jussà

4. 4
Outline
1. Motivation
2. Image Captioning
3. Visual Question Answering / Reasoning
4. Joint Embeddings

5. 5
Outline
1. Motivation
2. Image and Video Captioning
3. Visual Question Answering / Reasoning
4. Joint Embeddings

6. 6

7. 7[course site]

8. Text (English)
8
Text (French)
Neural Machine Translation

9. 9
Neural Machine Translation
Kyunghyun Cho, “Introduction to Neural Machine Translation with GPUs” (2015)
Representation or
Embedding

10. 10
Representation or
Embedding

11. 11
Representation or
Embedding
Encoder Decoder

12. bit.ly/InsightDL2018
#InsightDL2018
12
Encoder-Decoder
Front View Side View
Representation of the sentence
Kyunghyun Cho, “Introduction to Neural Machine Translation with GPUs” (2015)
Cho, Kyunghyun, Bart Van Merriënboer, Caglar Gulcehre, Dzmitry Bahdanau, Fethi Bougares, Holger Schwenk, and Yoshua
Bengio. "Learning phrase representations using RNN encoder-decoder for statistical machine translation." arXiv preprint
arXiv:1406.1078 (2014).

13. 13
Representation or
Embedding
Encoder

14. bit.ly/InsightDL2018
#InsightDL2018
14
Encoder in three steps
Kyunghyun Cho, “Introduction to Neural Machine Translation with GPUs” (2015)
(2)
(3) (1) One hot encoding
(2) Word embedding
(3) Sequence summarization

15. bit.ly/InsightDL2018
#InsightDL2018
15
(1) One hot encoding
cat: xT
= [1,0,0, ..., 0]
dog: xT
= [0,1,0, ..., 0]
.
.
house: xT
= [0,0,0, …,0,1,0,...,0]
.
.
.

16. bit.ly/InsightDL2018
#InsightDL2018
16
(2) Word embeddings
Figure: Christopher Olah, Visualizing Representations

17. bit.ly/InsightDL2018
#InsightDL2018
17
(3) Sequence summarization
Kyunghyun Cho, “Introduction to Neural Machine Translation with GPUs” (2015)

18. 18
Representation or
Embedding
Decoder

19. bit.ly/InsightDL2018
#InsightDL2018
19
Decoder
Kyunghyun Cho, “Introduction to Neural Machine Translation with GPUs” (2015)
The Recurrent State (zi
) of the decoder is determined by:
1) summary vector hT
2) previous output word ui-1
3) previous state zi-1
hT

20. bit.ly/InsightDL2018
#InsightDL2018
20
Decoder
Kyunghyun Cho, “Introduction to Neural Machine Translation with GPUs” (2015)
With zi
updated, we can compute a probability pi
for each word i as an
output of the RNN:
hT

21. bit.ly/InsightDL2018
#InsightDL2018
21
Decoder
Kyunghyun Cho, “Introduction to Neural Machine Translation with GPUs” (2015)
More words for the decoded sentence are generated until a <EOS> (End Of
Sentence) “word” is predicted.
EOS

22. 22
Outline
1. Motivation
2. Image and Video Captioning
3. Visual Question Answering / Reasoning
4. Joint Embeddings

23. 23
Representation or
Embedding
Encoder Decoder

24. 24
Captioning: Show & Tell
Vinyals, Oriol, Alexander Toshev, Samy Bengio, and Dumitru Erhan. "Show and tell: A neural image caption
generator." CVPR 2015. [video]

25. 25
Captioning: DeepImageSent
(Slides by Marc Bolaños): Karpathy, Andrej, and Li Fei-Fei. "Deep visual-semantic alignments for generating
image descriptions." CVPR 2015

26. 26
Captioning: DeepImageSent
(Slides by Marc Bolaños): Karpathy, Andrej, and Li Fei-Fei. "Deep visual-semantic alignments for
generating image descriptions." CVPR 2015
only takes into account
image features in the first
hidden state
Multimodal Recurrent
Neural Network

27. Challenge on Multimodal Image Translation:
http://www.statmt.org/wmt17/multimodal-task.html#task1
Multimodal Machine Translation

28. 28
Captioning: Show, Attend & Tell
Xu, Kelvin, Jimmy Ba, Ryan Kiros, Kyunghyun Cho, Aaron C. Courville, Ruslan Salakhutdinov, Richard S.
Zemel, and Yoshua Bengio. "Show, Attend and Tell: Neural Image Caption Generation with Visual
Attention." ICML 2015

29. 29
Captioning: Show, Attend & Tell
Xu, Kelvin, Jimmy Ba, Ryan Kiros, Kyunghyun Cho, Aaron C. Courville, Ruslan Salakhutdinov, Richard S.
Zemel, and Yoshua Bengio. "Show, Attend and Tell: Neural Image Caption Generation with Visual
Attention." ICML 2015

30. 30
Johnson, Justin, Andrej Karpathy, and Li Fei-Fei. "Densecap: Fully convolutional localization networks for
dense captioning." CVPR 2016
Captioning (+ Detection): DenseCap

31. 31
Captioning (+ Detection): DenseCap
Johnson, Justin, Andrej Karpathy, and Li Fei-Fei. "Densecap: Fully convolutional localization networks for
dense captioning." CVPR 2016

32. 32
Captioning (+ Detection): DenseCap
Johnson, Justin, Andrej Karpathy, and Li Fei-Fei. "Densecap: Fully convolutional localization networks for
dense captioning." CVPR 2016
XAVI: “man has
short hair”, “man
with short hair”
AMAIA:”a woman
wearing a black
shirt”, “
BOTH: “two men
wearing black
glasses”

33. 33
Captioning: Video
Jeffrey Donahue, Lisa Anne Hendricks, Sergio Guadarrama, Marcus Rohrbach, Subhashini Venugopalan,
Kate Saenko, Trevor Darrel. Long-term Recurrent Convolutional Networks for Visual Recognition and
Description, CVPR 2015. code

34. 34
(Slides by Marc Bolaños) Pingbo Pan, Zhongwen Xu, Yi Yang,Fei Wu,Yueting Zhuang Hierarchical
Recurrent Neural Encoder for Video Representation with Application to Captioning, CVPR 2016.
LSTM unit
(2nd layer)
Time
Image
t = 1 t = T
hidden state
at t = T
first chunk
of data
Captioning: Video

35. 35
Chung, Joon Son, Andrew Senior, Oriol Vinyals, and Andrew Zisserman. "Lip reading sentences in the wild."
CVPR 2017

36. 36
Lipreading: Watch, Listen, Attend & Spell
Chung, Joon Son, Andrew Senior, Oriol Vinyals, and Andrew Zisserman. "Lip reading sentences in the
wild." CVPR 2017
Audio
features
Image
features

37. 37
Lipreading: Watch, Listen, Attend & Spell
Chung, Joon Son, Andrew Senior, Oriol Vinyals, and Andrew Zisserman. "Lip reading sentences in the
wild." CVPR 2017
Attention over output
states from audio and
video is computed at
each timestep

38. 38
Assael, Yannis M., Brendan Shillingford, Shimon Whiteson, and Nando de Freitas. "LipNet: End-to-End
Sentence-level Lipreading." (2016).
Lip Reading: LipNet
Input (video frames) and output (sentence) sequences are not
aligned

39. 39
Graves et al. Connectionist Temporal Classification: Labelling Unsegmented Sequence Data with
Recurrent Neural Networks. ICML 2006
Lip Reading: LipNet
CTC Loss: Connectionist temporal classification
● Avoiding the need for alignment between input and output sequence by predicting
an additional “_” blank word
● Before computing the loss, repeated words and blank tokens are removed
● “a _ a b _ ” == “_ a a _ _ b b” == “a a b”

40. 40
Lip Reading: LipNet
Assael et al. LipNet: Sentence-level Lipreading. arXiv Nov 2016

41. 41
Outline
1. Neural Machine Transaltion (no vision here !)
2. Image and Video Captioning
3. Visual Question Answering / Reasoning
4. Joint Embeddings

42. 42
Visual Question Answering (VQA)
[z1
, z2
, … zN
] [y1
, y2
, … yM
]
“Is economic growth decreasing ?”
“Yes”
Encode
Encode
Decode

43. 43
Antol, Stanislaw, Aishwarya Agrawal, Jiasen Lu, Margaret Mitchell, Dhruv Batra, C. Lawrence Zitnick, and
Devi Parikh. "VQA: Visual question answering." CVPR 2015.
Visual Question Answering (VQA)

44. 44
Extract visual
features
Embedding
Predict answerMerge
Question
What object is flying?
Answer
Kite
Slide credit: Issey Masuda
Visual Question Answering (VQA)

45. 45
Visual Question Answering (VQA)
Masuda, Issey, Santiago Pascual de la Puente, and Xavier Giro-i-Nieto. "Open-Ended Visual
Question-Answering." ETSETB UPC TelecomBCN (2016).
Image
Question
Answer

46. 46
Visual Question Answering (VQA)
Francisco Roldán, Issey Masuda, Santiago Pascual de la Puente, and Xavier Giro-i-Nieto.
"Visual Question-Answering 2.0." ETSETB UPC TelecomBCN (2017).

47. 47
Noh, H., Seo, P. H., & Han, B. Image question answering using convolutional neural network with
dynamic parameter prediction. CVPR 2016
Dynamic Parameter Prediction Network (DPPnet)
Visual Question Answering (VQA)

48. 48
Visual Question Answering: Dynamic
(Slides and Slidecast by Santi Pascual): Xiong, Caiming, Stephen Merity, and Richard Socher. "Dynamic
Memory Networks for Visual and Textual Question Answering." ICML 2016

49. 49
Visual Question Answering: Grounded
(Slides and Screencast by Issey Masuda): Zhu, Yuke, Oliver Groth, Michael Bernstein, and Li Fei-Fei."Visual7W: Grounded
Question Answering in Images." CVPR 2016.

50. 50
Visual Dialog (Image Guessing Game)
Das, Abhishek, Satwik Kottur, Khushi Gupta, Avi Singh, Deshraj Yadav, José MF Moura, Devi Parikh, and Dhruv Batra.
"Visual Dialog." CVPR 2017

51. 51
Visual Dialog (Image Guessing Game)
Das, Abhishek, Satwik Kottur, Khushi Gupta, Avi Singh, Deshraj Yadav, José MF Moura, Devi Parikh, and Dhruv Batra.
"Visual Dialog." CVPR 2017

52. 52
Visual Reasoning
Johnson, Justin, Bharath Hariharan, Laurens van der Maaten, Li Fei-Fei, C. Lawrence Zitnick, and Ross
Girshick. "CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning."
CVPR 2017

53. 53
Visual Reasoning
(Slides by Fran Roldan) Justin Johnson, Bharath Hariharan, Laurens van der Maaten, Judy Hoffman, Fei-Fei Li, Larry
Zitnick, Ross Girshick , “Inferring and Executing Programs for Visual Reasoning”. ICCV 2017
Program Generator Execution Engine

54. 54
Outline
1. Neural Machine Transaltion (no vision here !)
2. Image and Video Captioning
3. Visual Question Answering / Reasoning
4. Joint Embeddings

55. 55
Representation or
EmbeddingEncoder
Joint Neural Embeddings
Encoder

56. 56
Frome, Andrea, Greg S. Corrado, Jon Shlens, Samy Bengio, Jeff Dean, and Tomas Mikolov. "Devise: A
deep visual-semantic embedding model." NIPS 2013
Joint Neural Embeddings

57. 57
Socher, R., Ganjoo, M., Manning, C. D., & Ng, A., Zero-shot learning through cross-modal transfer.
NIPS 2013 [slides] [code]
Zero-shot learning:
a class not present in the
training set of images
can be predicted
(eg. no images from
“cat” in the training set)
Joint Neural Embeddings

58. 58
Alejandro Woodward, Víctor Campos, Dèlia Fernàndez, Jordi Torres, Xavier Giró-i-Nieto,
Brendan Jou and Shih-Fu Chang (submitted)
Joint Neural Embeddings
Foggy Day

59. 59

60. 60
Amaia Salvador, Nicholas Haynes, Yusuf Aytar, Javier Marín, Ferda Ofli, Ingmar Weber,
Antonio Torralba, “Learning Cross-modal Embeddings for Cooking Recipes and Food
Images”. CVPR 2017
Image and text retrieval with joint embeddings.
Joint Neural Embeddings

61. 61
Amaia Salvador, Nicholas Haynes, Yusuf Aytar, Javier Marín, Ferda Ofli, Ingmar Weber,
Antonio Torralba, “Learning Cross-modal Embeddings for Cooking Recipes and Food
Images”. CVPR 2017
Joint Neural Embeddings

62. 62
Amaia Salvador, Nicholas Haynes, Yusuf Aytar, Javier Marín, Ferda Ofli, Ingmar Weber,
Antonio Torralba, “Learning Cross-modal Embeddings for Cooking Recipes and Food
Images”. CVPR 2017
Joint Neural Embeddings
joint
embedding
LSTM Bidirectional LSTM

63. 63
Image to image and text
Aytar, Yusuf, Lluis Castrejon, Carl Vondrick, Hamed Pirsiavash, and Antonio Torralba. "Cross-Modal
Scene Networks." CVPR 2016.

64. 64
Image to image and text
Aytar, Yusuf, Lluis Castrejon, Carl Vondrick, Hamed Pirsiavash, and Antonio Torralba. "Cross-Modal
Scene Networks." CVPR 2016.

65. Gella, Spandana, Rico Sennrich, Frank Keller, and Mirella Lapata. "Image Pivoting for Learning Multilingual Multimodal Representations." arXiv preprint
arXiv:1707.07601 (2017).
Janarthanan Rajendran, Mitesh M Khapra, Sarath Chandar, Balaraman Ravindran, Bridge Correlational Neural Networks for Multilingual Multimodal
Representation Learning NAACL, 2016
Multilingual & Multimodal Embeddings

66. 66
Outline
1. Motivation
2. Image Captioning
3. Visual Question Answering / Reasoning
4. Joint Embeddings

67. 67
Questions ?