Abstract | ||
---|---|---|
In an object recognition scenario with tens of thousands of categories, even a small number of labels per category leads to a very large number of total labels required. We propose a simple method of label sharing between semantically similar categories. We leverage the WordNet hierarchy to define semantic distance between any two categories and use this semantic distance to share labels. Our approach can be used with any classifier. Experimental results on a range of datasets, upto 80 million images and 75,000 categories in size, show that despite the simplicity of the approach, it leads to significant improvements in performance. |
Year | DOI | Venue |
---|---|---|
2010 | 10.1007/978-3-642-15549-9_55 | ECCV |
Keywords | Field | DocType |
Semantic Distance,Semantic Label,Training Pair,Internet Search Engine,Search Engine Result | Small number,Semantic similarity,Computer science,Large numbers,Artificial intelligence,Classifier (linguistics),Hierarchy,WordNet,Machine learning,Cognitive neuroscience of visual object recognition | Conference |
Volume | ISSN | ISBN |
6311 | 0302-9743 | 3-642-15548-0 |
Citations | PageRank | References |
59 | 2.09 | 19 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Robert Fergus | 1 | 11214 | 735.18 |
Hector Bernal | 2 | 59 | 2.09 |
Yair Weiss | 3 | 10240 | 834.60 |
Antonio Torralba | 4 | 14607 | 956.27 |