Improving Image Captioning by Leveraging Knowledge Graphs
Authors:
- Yimin ZhouYZYimin Zhou
- Yiwei SunYSYiwei Sun
- Vasant HonavarVHVasant Honavar
- 1 January 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 283-293
Abstract
We explore the use of a knowledge graphs, that capture general or commonsense knowledge, to augment the information extracted from images by the state-of-the-art methods for image captioning. We compare the performance of image captioning systems that as measured by CIDEr-D, a performance measure that is explicitly designed for evaluating image captioning systems, on several benchmark data sets such as MS COCO. The results of our experiments show that the variants of the state-of-the-art methods for image captioning that make use of the information extracted from knowledge graphs can substantially outperform those that rely solely on the information extracted from images.Keywords
All Related Versions
This publication has 24 references indexed in Scilit:
- Long-term recurrent convolutional networks for visual recognition and description2015
- Going deeper with convolutions2015
- Show and tell: A neural image caption generator2015
- CIDEr: Consensus-based image description evaluation2015
- Retrofitting Word Vectors to Semantic Lexicons2015
- From Image Annotation to Image Description2012
- Every Picture Tells a Story: Generating Sentences from Images2010
- The Pascal Visual Object Classes (VOC) ChallengeInternational Journal of Computer Vision, 2009
- What do we perceive in a glance of a real-world scene?Journal of Vision, 2007
- ConceptNet — A Practical Commonsense Reasoning Tool-KitBT Technology Journal, 2004