Cross-domain image description generation using transfer learning

Philip Kinghorn, Li Zhang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-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 languageEnglish
Title of host publicationData Science and Knowledge Engineering for Sensing Decision Support
Subtitle of host publicationProceedings of the 13th International FLINS Conference (FLINS 2018)
Place of PublicationNorthern Ireland
Pages1462-1469
Volume11
ISBN (Electronic)978-981-3273-24-5
DOIs
Publication statusPublished - Oct 2018

Fingerprint Dive into the research topics of 'Cross-domain image description generation using transfer learning'. Together they form a unique fingerprint.

Cite this