Abstract

With the progress in the field of computer vision, we are moving closer and closer towards the ultimate aim of human like vision for machines. Scene understanding is an essential part of this research. It seeks the goal that any image should be as understandable and decipherable for computers as it is for humans. The stall in the progress of the different components of scene understanding, due to the limitations of the traditional algorithms, has now been broken by the induction of neural networks for computer vision tasks. The advancements in parallel computational hardware has made it possible to train very deep and complex neural network architectures. This has vastly improved the performances of algorithms for all the different components of scene understanding. This chapter analyses these contributions of deep learning and also presents the advancements of high level scene understanding tasks, such as caption generation for images. It also sheds light on the need to combine these individual components into an integrated system.
Original languageEnglish
Title of host publicationHandbook of deep learning applications
EditorsValentina Emilia Balas, Sanjiban Sekhar Roy, Dharmendra Sharma, Pijush Samui
Place of PublicationCham
PublisherSpringer
Pages21-51
Number of pages27
ISBN (Electronic)9783030114794
ISBN (Print)9783030114787
DOIs
Publication statusPublished - 19 Mar 2019

Publication series

NameSmart Innovation, Systems and Technologies
PublisherSpringer International Publishing
Volume136
ISSN (Electronic)2190-3018

Fingerprint

Computer vision
Neural networks
Network architecture
Hardware
Deep learning

Cite this

Nadeem, U., Shah, S. A. A., Sohel, F., Togneri, R., & Bennamoun, M. (2019). Deep learning for scene understanding. In V. E. Balas, S. S. Roy, D. Sharma, & P. Samui (Eds.), Handbook of deep learning applications (pp. 21-51). (Smart Innovation, Systems and Technologies; Vol. 136). Cham: Springer. https://doi.org/10.1007/978-3-030-11479-4_2
Nadeem, Uzair ; Shah, Syed Afaq Ali ; Sohel, Ferdous ; Togneri, Roberto ; Bennamoun, Mohammed. / Deep learning for scene understanding. Handbook of deep learning applications. editor / Valentina Emilia Balas ; Sanjiban Sekhar Roy ; Dharmendra Sharma ; Pijush Samui. Cham : Springer, 2019. pp. 21-51 (Smart Innovation, Systems and Technologies).
@inbook{95f123b3a422437bb5aa6caccbd771e1,
title = "Deep learning for scene understanding",
abstract = "With the progress in the field of computer vision, we are moving closer and closer towards the ultimate aim of human like vision for machines. Scene understanding is an essential part of this research. It seeks the goal that any image should be as understandable and decipherable for computers as it is for humans. The stall in the progress of the different components of scene understanding, due to the limitations of the traditional algorithms, has now been broken by the induction of neural networks for computer vision tasks. The advancements in parallel computational hardware has made it possible to train very deep and complex neural network architectures. This has vastly improved the performances of algorithms for all the different components of scene understanding. This chapter analyses these contributions of deep learning and also presents the advancements of high level scene understanding tasks, such as caption generation for images. It also sheds light on the need to combine these individual components into an integrated system.",
author = "Uzair Nadeem and Shah, {Syed Afaq Ali} and Ferdous Sohel and Roberto Togneri and Mohammed Bennamoun",
year = "2019",
month = "3",
day = "19",
doi = "10.1007/978-3-030-11479-4_2",
language = "English",
isbn = "9783030114787",
series = "Smart Innovation, Systems and Technologies",
publisher = "Springer",
pages = "21--51",
editor = "Balas, {Valentina Emilia } and Roy, {Sanjiban Sekhar } and Sharma, {Dharmendra } and Samui, {Pijush }",
booktitle = "Handbook of deep learning applications",
address = "Netherlands",

}

Nadeem, U, Shah, SAA, Sohel, F, Togneri, R & Bennamoun, M 2019, Deep learning for scene understanding. in VE Balas, SS Roy, D Sharma & P Samui (eds), Handbook of deep learning applications. Smart Innovation, Systems and Technologies, vol. 136, Springer, Cham, pp. 21-51. https://doi.org/10.1007/978-3-030-11479-4_2

Deep learning for scene understanding. / Nadeem, Uzair; Shah, Syed Afaq Ali; Sohel, Ferdous; Togneri, Roberto; Bennamoun, Mohammed.

Handbook of deep learning applications. ed. / Valentina Emilia Balas; Sanjiban Sekhar Roy; Dharmendra Sharma; Pijush Samui. Cham : Springer, 2019. p. 21-51 (Smart Innovation, Systems and Technologies; Vol. 136).

Research output: Chapter in Book/Conference paperChapter

TY - CHAP

T1 - Deep learning for scene understanding

AU - Nadeem, Uzair

AU - Shah, Syed Afaq Ali

AU - Sohel, Ferdous

AU - Togneri, Roberto

AU - Bennamoun, Mohammed

PY - 2019/3/19

Y1 - 2019/3/19

N2 - With the progress in the field of computer vision, we are moving closer and closer towards the ultimate aim of human like vision for machines. Scene understanding is an essential part of this research. It seeks the goal that any image should be as understandable and decipherable for computers as it is for humans. The stall in the progress of the different components of scene understanding, due to the limitations of the traditional algorithms, has now been broken by the induction of neural networks for computer vision tasks. The advancements in parallel computational hardware has made it possible to train very deep and complex neural network architectures. This has vastly improved the performances of algorithms for all the different components of scene understanding. This chapter analyses these contributions of deep learning and also presents the advancements of high level scene understanding tasks, such as caption generation for images. It also sheds light on the need to combine these individual components into an integrated system.

AB - With the progress in the field of computer vision, we are moving closer and closer towards the ultimate aim of human like vision for machines. Scene understanding is an essential part of this research. It seeks the goal that any image should be as understandable and decipherable for computers as it is for humans. The stall in the progress of the different components of scene understanding, due to the limitations of the traditional algorithms, has now been broken by the induction of neural networks for computer vision tasks. The advancements in parallel computational hardware has made it possible to train very deep and complex neural network architectures. This has vastly improved the performances of algorithms for all the different components of scene understanding. This chapter analyses these contributions of deep learning and also presents the advancements of high level scene understanding tasks, such as caption generation for images. It also sheds light on the need to combine these individual components into an integrated system.

U2 - 10.1007/978-3-030-11479-4_2

DO - 10.1007/978-3-030-11479-4_2

M3 - Chapter

SN - 9783030114787

T3 - Smart Innovation, Systems and Technologies

SP - 21

EP - 51

BT - Handbook of deep learning applications

A2 - Balas, Valentina Emilia

A2 - Roy, Sanjiban Sekhar

A2 - Sharma, Dharmendra

A2 - Samui, Pijush

PB - Springer

CY - Cham

ER -

Nadeem U, Shah SAA, Sohel F, Togneri R, Bennamoun M. Deep learning for scene understanding. In Balas VE, Roy SS, Sharma D, Samui P, editors, Handbook of deep learning applications. Cham: Springer. 2019. p. 21-51. (Smart Innovation, Systems and Technologies). https://doi.org/10.1007/978-3-030-11479-4_2