Abstract | ||
---|---|---|
We address the problem of large-scale fine-grained visual categorization, describing new methods we have used to produce an online field guide to 500 North American bird species. We focus on the challenges raised when such a system is asked to distinguish between highly similar species of birds. First, we introduce \"one-vs-most classifiers.\" By eliminating highly similar species during training, these classifiers achieve more accurate and intuitive results than common one-vs-all classifiers. Second, we show how to estimate spatio-temporal class priors from observations that are sampled at irregular and biased locations. We show how these priors can be used to significantly improve performance. We then show state-of-the-art recognition performance on a new, large dataset that we make publicly available. These recognition methods are integrated into the online field guide, which is also publicly available. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1109/CVPR.2014.259 | CVPR |
Keywords | Field | DocType |
image classification,North American bird species,birdsnap,large-scale fine-grained visual categorization,one-vs-most classifiers,recognition performance,spatio-temporal class estimation,Fine-grained visual categorization,birds,large-scale classification,recognition,species identification | Computer vision,Categorization,Pattern recognition,Computer science,Species identification,Artificial intelligence,Contextual image classification,Prior probability,Machine learning | Conference |
ISSN | Citations | PageRank |
1063-6919 | 57 | 1.55 |
References | Authors | |
20 | 6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Thomas Berg | 1 | 57 | 1.55 |
Jiongxin Liu | 2 | 158 | 6.34 |
Seung Woo Lee | 3 | 57 | 1.55 |
Michelle L. Alexander | 4 | 57 | 1.55 |
David W. Jacobs | 5 | 4599 | 348.03 |
Peter N. Belhumeur | 6 | 12242 | 1001.27 |