Flickr-tag prediction using multi-modal fusion and meta information
Journal
MM 2013 - Proceedings of the 2013 ACM Multimedia Conference
Pages
353-356
ISBN
9781450324045
Date Issued
2013
Author(s)
Abstract
We present our evaluation and analysis on Yahoo Largescale Flickr-tag Image Classification dataset. Our evaluations show that combining multi-features and different classification models, the MAP of tag prediction can be significantly improve over ordinary linear classification. Further analysis shows that some tags are given not because of the visual content but the meta information of images. Our experiments show that we can make more accurate prediction on certain tags using meta information without any training process, compared with visual content based classifiers. Combine the meta information, multi-features and multimodels fusion, we achieve significantly better performance than simple linear classification. We also evaluate the performance of various mid-level feature, and the results suggest that "Concept Bank" feature may be a promising direction for the task. Copyright ? 2013 ACM.
Subjects
Concept bank; Meta data; Multi-features fusion
Other Subjects
Accurate prediction; Better performance; Classification models; Concept bank; Evaluation and analysis; Linear classification; Mid-level features; Multi-features fusions; Metadata; Image classification
Type
conference paper