GCap: Graph-based automatic image captioning
Author
Pan J.-Y.
Yang H.-J.
Faloutsos C.
Duygulu, Pınar
Date
2004Source Title
2004 Conference on Computer Vision and Pattern Recognition Workshop
Print ISSN
2160-7508
Publisher
IEEE
Language
English
Type
Conference PaperItem Usage Stats
151
views
views
268
downloads
downloads
Abstract
Given an image, how do we automatically assign keywords to it? In this paper, we propose a novel, graph-based approach (GCap) which outperforms previously reported methods for automatic image captioning. Moreover, it is fast and scales well, with its training and testing time linear to the data set size. We report auto-captioning experiments on the "standard" Corel image database of 680 MBytes, where GCap outperforms recent, successful auto-captioning methods by up to 10 percentage points in captioning accuracy (50% relative improvement). © 2004 IEEE.
Keywords
Computer visionImage retrieval
Pattern recognition
Statistical tests
Automatic image captioning
Corel image database
Data set size
Graph-based
Percentage points
Training and testing
Graphic methods