Constrained metric learning by permutation inducing isometries
المؤلف | Bosveld, Joel |
المؤلف | Mahmood, Arif |
المؤلف | Huynh, Du Q. |
المؤلف | Noakes, Lyle |
تاريخ الإتاحة | 2021-09-05T05:40:15Z |
تاريخ النشر | 2016 |
اسم المنشور | IEEE Transactions on Image Processing |
المصدر | Scopus |
الرقم المعياري الدولي للكتاب | 10577149 |
الملخص | The choice of metric critically affects the performance of classification and clustering algorithms. Metric learning algorithms attempt to improve performance, by learning a more appropriate metric. Unfortunately, most of the current algorithms learn a distance function which is not invariant to rigid transformations of images. Therefore, the distances between two images and their rigidly transformed pair may differ, leading to inconsistent classification or clustering results. We propose to constrain the learned metric to be invariant to the geometry preserving transformations of images that induce permutations in the feature space. The constraint that these transformations are isometries of the metric ensures consistent results and improves accuracy. Our second contribution is a dimension reduction technique that is consistent with the isometry constraints. Our third contribution is the formulation of the isometry constrained logistic discriminant metric learning (IC-LDML) algorithm, by incorporating the isometry constraints within the objective function of the LDML algorithm. The proposed algorithm is compared with the existing techniques on the publicly available labeled faces in the wild, viewpoint-invariant pedestrian recognition, and Toy Cars data sets. The IC-LDML algorithm has outperformed existing techniques for the tasks of face recognition, person identification, and object classification by a significant margin. |
اللغة | en |
الناشر | Institute of Electrical and Electronics Engineers Inc. |
الموضوع | Constrained metric learning Face recognition Isometry constraints Logistic-discriminant metric learning Object classification Pedestrian identification |
النوع | Article |
الصفحات | 92-103 |
رقم العدد | 1 |
رقم المجلد | 25 |
الملفات في هذه التسجيلة
الملفات | الحجم | الصيغة | العرض |
---|---|---|---|
لا توجد ملفات لها صلة بهذه التسجيلة. |
هذه التسجيلة تظهر في المجموعات التالية
-
علوم وهندسة الحاسب [2288 items ]