Natural language image descriptor

Anurag Kishore; Sanjay Singh

doi:10.1109/RAICS.2015.7488398

Natural language image descriptor

Anurag Kishore, Sanjay Singh

Department of Information and Communication Technology, Manipal Institute of Technology, Manipal

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Generating descriptions for visual data (images and video) automatically has been a complicated task in the field of Computer Vision and Artificial Intelligence. This paper discusses the working of and improvements on an algorithm called Neural Image Captioner (NIC) by Oriol Vinyals and his team, which uses a deep convolutional and recurrent architecture to generate natural language sentences to describe the visual data input. We look at the possibility of making this algorithm train faster without allowing it to lose accuracy via the usage of techniques like Stochastic Gradient Descent and also employ an algorithm to find the perfect depth of the convolutional part of the network for different datasets. A drop of 33% was observed in the number of iterations required to get the algorithm to its original proficiency as claimed by Oriol et al.

Original language	English
Title of host publication	2015 IEEE Recent Advances in Intelligent Computational Systems, RAICS 2015
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	110-115
Number of pages	6
ISBN (Electronic)	9781467366700
DOIs	https://doi.org/10.1109/RAICS.2015.7488398
Publication status	Published - 09-06-2016
Event	2015 IEEE Recent Advances in Intelligent Computational Systems, RAICS 2015 - Trivandrum, Kerala, India Duration: 10-12-2015 → 12-12-2015

Conference

Conference	2015 IEEE Recent Advances in Intelligent Computational Systems, RAICS 2015
Country/Territory	India
City	Trivandrum, Kerala
Period	10-12-15 → 12-12-15

All Science Journal Classification (ASJC) codes

Artificial Intelligence
Computer Science Applications
Control and Systems Engineering

Access to Document

10.1109/RAICS.2015.7488398

Cite this

@inproceedings{b96f939bead8450389bdc2a61961f791,

title = "Natural language image descriptor",

abstract = "Generating descriptions for visual data (images and video) automatically has been a complicated task in the field of Computer Vision and Artificial Intelligence. This paper discusses the working of and improvements on an algorithm called Neural Image Captioner (NIC) by Oriol Vinyals and his team, which uses a deep convolutional and recurrent architecture to generate natural language sentences to describe the visual data input. We look at the possibility of making this algorithm train faster without allowing it to lose accuracy via the usage of techniques like Stochastic Gradient Descent and also employ an algorithm to find the perfect depth of the convolutional part of the network for different datasets. A drop of 33% was observed in the number of iterations required to get the algorithm to its original proficiency as claimed by Oriol et al.",

author = "Anurag Kishore and Sanjay Singh",

year = "2016",

month = jun,

day = "9",

doi = "10.1109/RAICS.2015.7488398",

language = "English",

pages = "110--115",

booktitle = "2015 IEEE Recent Advances in Intelligent Computational Systems, RAICS 2015",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

address = "United States",

note = "2015 IEEE Recent Advances in Intelligent Computational Systems, RAICS 2015 ; Conference date: 10-12-2015 Through 12-12-2015",

}

Kishore, A & Singh, S 2016, Natural language image descriptor. in 2015 IEEE Recent Advances in Intelligent Computational Systems, RAICS 2015., 7488398, Institute of Electrical and Electronics Engineers Inc., pp. 110-115, 2015 IEEE Recent Advances in Intelligent Computational Systems, RAICS 2015, Trivandrum, Kerala, India, 10-12-15. https://doi.org/10.1109/RAICS.2015.7488398

TY - GEN

T1 - Natural language image descriptor

AU - Kishore, Anurag

AU - Singh, Sanjay

PY - 2016/6/9

Y1 - 2016/6/9

N2 - Generating descriptions for visual data (images and video) automatically has been a complicated task in the field of Computer Vision and Artificial Intelligence. This paper discusses the working of and improvements on an algorithm called Neural Image Captioner (NIC) by Oriol Vinyals and his team, which uses a deep convolutional and recurrent architecture to generate natural language sentences to describe the visual data input. We look at the possibility of making this algorithm train faster without allowing it to lose accuracy via the usage of techniques like Stochastic Gradient Descent and also employ an algorithm to find the perfect depth of the convolutional part of the network for different datasets. A drop of 33% was observed in the number of iterations required to get the algorithm to its original proficiency as claimed by Oriol et al.

AB - Generating descriptions for visual data (images and video) automatically has been a complicated task in the field of Computer Vision and Artificial Intelligence. This paper discusses the working of and improvements on an algorithm called Neural Image Captioner (NIC) by Oriol Vinyals and his team, which uses a deep convolutional and recurrent architecture to generate natural language sentences to describe the visual data input. We look at the possibility of making this algorithm train faster without allowing it to lose accuracy via the usage of techniques like Stochastic Gradient Descent and also employ an algorithm to find the perfect depth of the convolutional part of the network for different datasets. A drop of 33% was observed in the number of iterations required to get the algorithm to its original proficiency as claimed by Oriol et al.

UR - http://www.scopus.com/inward/record.url?scp=84979009562&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84979009562&partnerID=8YFLogxK

U2 - 10.1109/RAICS.2015.7488398

DO - 10.1109/RAICS.2015.7488398

M3 - Conference contribution

AN - SCOPUS:84979009562

SP - 110

EP - 115

BT - 2015 IEEE Recent Advances in Intelligent Computational Systems, RAICS 2015

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2015 IEEE Recent Advances in Intelligent Computational Systems, RAICS 2015

Y2 - 10 December 2015 through 12 December 2015

ER -

Natural language image descriptor

Abstract

Conference

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this