The Tangent Earth Mover's Distance
نویسندگان
چکیده
We present a new histogram distance, the Tangent Earth Mover’s Distance (TEMD). The TEMD is a generalization of the Earth Mover’s Distance (EMD) that is invariant to some global transformations. Thus, like the EMD it is robust to local deformations. Additionally, it is robuster to global transformations such as global translations and rotations of the whole image. The TEMD is formulated as a linear program which allows efficient computation. Additionally, previous works about the efficient computation of the EMD that reduced the number of variables in the EMD linear program can be used to accelerate also the TEMD computation. We present results for image retrieval using the Scale Invariant Feature Transform (SIFT) and color image descriptors. We show that the new TEMD outperforms state of the art distances.
منابع مشابه
Earth Mover's Distance and Equivalent Metrics for Spaces with Semigroups
introduce a multi-scale metric on a space equipped with a diffusion semigroup. We prove, under some technical conditions, that the norm dual to the space of Lipschitz functions with respect to this metric is equivalent to two other norms, one of which is a weighted sum of the averages at each scale, and one of which is a weighted sum of the difference of averages across scales. The notion of 's...
متن کاملMetric-Preserving Reduction of Earth Mover's Distance
We prove that the earth mover’s distance problem reduces to a problem with half the number of constraints regardless of the ground distance, and propose a further reduced formulation when the ground distance comes from a graph with a homogeneous neighborhood structure. We also propose to apply our formulation to the non-negative matrix factorization.
متن کاملSupervised Earth Mover's Distance Learning and Its Computer Vision Applications
Earth Mover’s Distance (EMD) is an intuitive and natural distance metric for comparing two histograms or probability distributions. We propose to jointly optimize the ground distance matrix and the EMD flow-network based on partial ordering of histogram distances in an optimization framework. Two applications in computer vision are used to demonstrate the effectiveness of the algorithm: firstly...
متن کاملSpatial color image retrieval without segmentation using thumbnails and the Earth Mover's Distance
We introduce a spatial color image retrieval method which does not include any segmentation step. This method relies on small image thumbnails and the Earth Mover’s Distance (EMD). We then derive an unsupervised matching criterion using an a contrario approach. Experiments are performed on a database of illuminated manuscripts.
متن کاملInducing Bilingual Lexica From Non-Parallel Data With Earth Mover's Distance Regularization
pages 3188–3198, Osaka, Japan, December 11-17 2016. Inducing Bilingual Lexica From Non-Parallel Data With Earth Mover’s Distance Regularization Meng Zhang†‡ Yang Liu†‡ Huanbo Luan† Yiqun Liu† Maosong Sun†‡ †State Key Laboratory of Intelligent Technology and Systems Tsinghua National Laboratory for Information Science and Technology Department of Computer Science and Technology, Tsinghua Univers...
متن کامل