Decremental Generalized Discriminative Common Vectors applied to images classification

Katerine Diaz-Chito, Jesus Martinez del Rincon, Aura Hernandez-Sabate

Research output: Contribution to journalArticlepeer-review

2 Citations (Scopus)
184 Downloads (Pure)


In this paper, a novel decremental subspace-based learning method called DecrementalGeneralized Discriminative Common Vectors method (DGDCV) is presented.The method makes use of the concept of decremental learning, whichwe introduce in the field of supervised feature extraction and classification. Byefficiently removing unnecessary data and/or classes for a knowledge base, ourmethodology is able to update the model without recalculating the full projectionor accessing to the previously processed training data, while retaining thepreviously acquired knowledge. The proposed method has been validated in 6standard face recognition datasets, showing a considerable computational gainwithout compromising the accuracy of the model.
Original languageEnglish
Pages (from-to)46-57
JournalKnowledge-Based Systems
Early online date25 May 2017
Publication statusPublished - 01 Sep 2017


Dive into the research topics of 'Decremental Generalized Discriminative Common Vectors applied to images classification'. Together they form a unique fingerprint.

Cite this