Comparing data-dependent and data-independent embeddings for classification and ranking of Internet images

This paper presents a comparative evaluation of feature embeddings for classification and ranking in large-scale Internet image datasets. We follow a popular framework for scalable visual learning, in which the data is first transformed by a nonlinear embedding and then an efficient linear classifier is trained in the resulting space. Our study includes data-dependent embeddings inspired by the semi-supervised learning literature, and data-independent ones based on approximating specific kernels (such as the Gaussian kernel for GIST features and the histogram intersection kernel for bags of words). Perhaps surprisingly, we find that data-dependent embeddings, despite being computed from large amounts of unlabeled data, do not have any advantage over data-independent ones in the regime of scarce labeled data. On the other hand, we find that several data-dependent embeddings are competitive with popular data-independent choices for large-scale classification.