Cross-domain image description generation using transfer learning

Philip Kinghorn; Li Zhang

doi:10.1142/9789813273238_0182

Cross-domain image description generation using transfer learning

Philip Kinghorn, Li Zhang

School of Computer Science

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

Abstract

In this research, we present image description generation using deep and transfer learning techniques for cross-domain images. The proposed system is able to conduct image captioning for out-of-scope images via fine-tuning and transfer learning. Tested with several cross-domain datasets, the work shows impressive capabilities for describing out-of-domain CCTV surveillance, artwork and healthcare images.

Original language	English
Title of host publication	Data Science and Knowledge Engineering for Sensing Decision Support
Subtitle of host publication	Proceedings of the 13th International FLINS Conference (FLINS 2018)
Place of Publication	Northern Ireland
Pages	1462-1469
Volume	11
ISBN (Electronic)	978-981-3273-24-5
DOIs	https://doi.org/10.1142/9789813273238_0182
Publication status	Published - Oct 2018

Keywords

Deep learning
Transfer learning
Image description generation

Access to Document

10.1142/9789813273238_0182

Cite this

@inproceedings{9e2205a33d984617b278d76b83e09b99,

title = "Cross-domain image description generation using transfer learning",

abstract = "In this research, we present image description generation using deep and transfer learning techniques for cross-domain images. The proposed system is able to conduct image captioning for out-of-scope images via fine-tuning and transfer learning. Tested with several cross-domain datasets, the work shows impressive capabilities for describing out-of-domain CCTV surveillance, artwork and healthcare images. ",

keywords = "Deep learning, Transfer learning, Image description generation",

author = "Philip Kinghorn and Li Zhang",

year = "2018",

month = oct,

doi = "10.1142/9789813273238\_0182",

language = "English",

isbn = "978-981-3273-22-1",

volume = "11",

pages = "1462--1469",

booktitle = "Data Science and Knowledge Engineering for Sensing Decision Support",

}

Cross-domain image description generation using transfer learning. / Kinghorn, Philip; Zhang, Li.
Data Science and Knowledge Engineering for Sensing Decision Support: Proceedings of the 13th International FLINS Conference (FLINS 2018). Vol. 11 Northern Ireland, 2018. p. 1462-1469.

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

TY - GEN

T1 - Cross-domain image description generation using transfer learning

AU - Kinghorn, Philip

AU - Zhang, Li

PY - 2018/10

Y1 - 2018/10

N2 - In this research, we present image description generation using deep and transfer learning techniques for cross-domain images. The proposed system is able to conduct image captioning for out-of-scope images via fine-tuning and transfer learning. Tested with several cross-domain datasets, the work shows impressive capabilities for describing out-of-domain CCTV surveillance, artwork and healthcare images.

AB - In this research, we present image description generation using deep and transfer learning techniques for cross-domain images. The proposed system is able to conduct image captioning for out-of-scope images via fine-tuning and transfer learning. Tested with several cross-domain datasets, the work shows impressive capabilities for describing out-of-domain CCTV surveillance, artwork and healthcare images.

KW - Deep learning

KW - Transfer learning

KW - Image description generation

U2 - 10.1142/9789813273238_0182

DO - 10.1142/9789813273238_0182

M3 - Conference contribution

SN - 978-981-3273-22-1

VL - 11

SP - 1462

EP - 1469

BT - Data Science and Knowledge Engineering for Sensing Decision Support

CY - Northern Ireland

ER -

Cross-domain image description generation using transfer learning

Abstract

Keywords

Access to Document

Fingerprint

Cite this