English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Learning a fast and discriminative patch descriptor is a challenging topic incomputer vision. Recently, many existing works focus on training variousdescriptor learning networks by minimizing a triplet loss (or its variants),which is expected to decrease the distance between each positive pair andincrease the distance between each negative pair. However, such an expectationhas to be lowered due to the non-perfect convergence of network optimizer to alocal solution. Addressing this problem and the open computational speedproblem, we propose a Descriptor Distillation framework for local descriptorlearning, called DesDis, where a student model gains knowledge from apre-trained teacher model, and it is further enhanced via a designedteacher-student regularizer. This teacher-student regularizer is to constrainthe difference between the positive (also negative) pair similarity from theteacher model and that from the student model, and we theoretically prove thata more effective student model could be trained by minimizing a weightedcombination of the triplet loss and this regularizer, than its teacher which istrained by minimizing the triplet loss singly. Under the proposed DesDis, manyexisting descriptor networks could be embedded as the teacher model, andaccordingly, both equal-weight and light-weight student models could bederived, which outperform their teacher in either accuracy or speed.Experimental results on 3 public datasets demonstrate that the equal-weightstudent models, derived from the proposed DesDis framework by utilizing threetypical descriptor learning networks as teacher models, could achievesignificantly better performances than their teachers and several othercomparative methods. In addition, the derived light-weight models could achieve8 times or even faster speeds than the comparative methods under similar patchverification performances

Descriptor Distillation: a Teacher-Student-Regularized Framework for Learning Local Descriptors