Face Recognition based on Discriminative Group Structured Dictionary
Available Online December 2018.
- https://doi.org/10.2991/tlicsc-18.2018.56How to use a DOI?
- Face recognition, discriminative group structured dictionary, group sparsity, linear discriminant analysis, low rank matrix recovery.
- Given a group structured dictionary, sparse representation vector is more likely to take on a particular structure. Intuitively, better results can be achieved if the structural characteristics are reasonably utilized. Inspired by this, we propose to construct a new discriminative group structured dictionary and impose a group sparse structure on the desired sparse representation vector. Thus, each testing sample can be decomposed as the product of the discriminative group structured dictionary and a group sparse representation vector. The identity of testing sample can be found by evaluating which class could induce the minimum reconstruction residual. Numerous experimental results on benchmark face databases demonstrate the effectiveness of the proposed method.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Yuhua Li PY - 2018/12 DA - 2018/12 TI - Face Recognition based on Discriminative Group Structured Dictionary BT - 2018 International Conference on Transportation & Logistics, Information & Communication, Smart City (TLICSC 2018) PB - Atlantis Press SP - 351 EP - 358 SN - 1951-6851 UR - https://doi.org/10.2991/tlicsc-18.2018.56 DO - https://doi.org/10.2991/tlicsc-18.2018.56 ID - Li2018/12 ER -