Matches in SemOpenAlex for { <https://semopenalex.org/work/W2510111515> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W2510111515 abstract "One task of heterogeneous face recognition is to match a near infrared (NIR) face image to a visible light (VIS) image. In practice, there are often a few pairwise NIR-VIS face images but it is easy to collect lots of VIS face images. Therefore, how to use these unpaired VIS images to improve the NIR-VIS recognition accuracy is an ongoing issue. This paper presents a deep TransfeR NIR-VIS heterogeneous facE recognition neTwork (TRIVET) for NIR-VIS face recognition. First, to utilize large numbers of unpaired VIS face images, we employ the deep convolutional neural network (CNN) with ordinal measures to learn discriminative models. The ordinal activation function (Max-Feature-Map) is used to select discriminative features and make the models robust and lighten. Second, we transfer these models to NIR-VIS domain by fine-tuning with two types of NIR-VIS triplet loss. The triplet loss not only reduces intra-class NIR-VIS variations but also augments the number of positive training sample pairs. It makes fine-tuning deep models on a small dataset possible. The proposed method achieves state-of-the-art recognition performance on the most challenging CASIA NIR-VIS 2.0 Face Database. It achieves a new record on rank-1 accuracy of 95.74% and verification rate of 91.03% at FAR=0.001. It cuts the error rate in comparison with the best accuracy [27] by 69%." @default.
- W2510111515 created "2016-09-16" @default.
- W2510111515 creator A5013444814 @default.
- W2510111515 creator A5049612204 @default.
- W2510111515 creator A5063589855 @default.
- W2510111515 creator A5078748099 @default.
- W2510111515 date "2016-06-01" @default.
- W2510111515 modified "2023-09-27" @default.
- W2510111515 title "Transferring deep representation for NIR-VIS heterogeneous face recognition" @default.
- W2510111515 cites W1607277929 @default.
- W2510111515 cites W1907941930 @default.
- W2510111515 cites W1913277147 @default.
- W2510111515 cites W1965996540 @default.
- W2510111515 cites W1976948919 @default.
- W2510111515 cites W2016957240 @default.
- W2510111515 cites W2051464482 @default.
- W2510111515 cites W2053251013 @default.
- W2510111515 cites W2085810284 @default.
- W2510111515 cites W2115403315 @default.
- W2510111515 cites W2121604810 @default.
- W2510111515 cites W2131216994 @default.
- W2510111515 cites W2194321275 @default.
- W2510111515 cites W2214409633 @default.
- W2510111515 cites W2325939864 @default.
- W2510111515 doi "https://doi.org/10.1109/icb.2016.7550064" @default.
- W2510111515 hasPublicationYear "2016" @default.
- W2510111515 type Work @default.
- W2510111515 sameAs 2510111515 @default.
- W2510111515 citedByCount "37" @default.
- W2510111515 countsByYear W25101115152015 @default.
- W2510111515 countsByYear W25101115152016 @default.
- W2510111515 countsByYear W25101115152017 @default.
- W2510111515 countsByYear W25101115152018 @default.
- W2510111515 countsByYear W25101115152019 @default.
- W2510111515 countsByYear W25101115152020 @default.
- W2510111515 countsByYear W25101115152021 @default.
- W2510111515 countsByYear W25101115152022 @default.
- W2510111515 countsByYear W25101115152023 @default.
- W2510111515 crossrefType "proceedings-article" @default.
- W2510111515 hasAuthorship W2510111515A5013444814 @default.
- W2510111515 hasAuthorship W2510111515A5049612204 @default.
- W2510111515 hasAuthorship W2510111515A5063589855 @default.
- W2510111515 hasAuthorship W2510111515A5078748099 @default.
- W2510111515 hasConcept C108583219 @default.
- W2510111515 hasConcept C138885662 @default.
- W2510111515 hasConcept C144024400 @default.
- W2510111515 hasConcept C153180895 @default.
- W2510111515 hasConcept C154945302 @default.
- W2510111515 hasConcept C184898388 @default.
- W2510111515 hasConcept C2776401178 @default.
- W2510111515 hasConcept C2779304628 @default.
- W2510111515 hasConcept C31510193 @default.
- W2510111515 hasConcept C36289849 @default.
- W2510111515 hasConcept C41008148 @default.
- W2510111515 hasConcept C41895202 @default.
- W2510111515 hasConcept C52622490 @default.
- W2510111515 hasConcept C81363708 @default.
- W2510111515 hasConcept C97931131 @default.
- W2510111515 hasConceptScore W2510111515C108583219 @default.
- W2510111515 hasConceptScore W2510111515C138885662 @default.
- W2510111515 hasConceptScore W2510111515C144024400 @default.
- W2510111515 hasConceptScore W2510111515C153180895 @default.
- W2510111515 hasConceptScore W2510111515C154945302 @default.
- W2510111515 hasConceptScore W2510111515C184898388 @default.
- W2510111515 hasConceptScore W2510111515C2776401178 @default.
- W2510111515 hasConceptScore W2510111515C2779304628 @default.
- W2510111515 hasConceptScore W2510111515C31510193 @default.
- W2510111515 hasConceptScore W2510111515C36289849 @default.
- W2510111515 hasConceptScore W2510111515C41008148 @default.
- W2510111515 hasConceptScore W2510111515C41895202 @default.
- W2510111515 hasConceptScore W2510111515C52622490 @default.
- W2510111515 hasConceptScore W2510111515C81363708 @default.
- W2510111515 hasConceptScore W2510111515C97931131 @default.
- W2510111515 hasLocation W25101115151 @default.
- W2510111515 hasOpenAccess W2510111515 @default.
- W2510111515 hasPrimaryLocation W25101115151 @default.
- W2510111515 hasRelatedWork W1775397219 @default.
- W2510111515 hasRelatedWork W2061273563 @default.
- W2510111515 hasRelatedWork W2136485282 @default.
- W2510111515 hasRelatedWork W2279398222 @default.
- W2510111515 hasRelatedWork W2285052147 @default.
- W2510111515 hasRelatedWork W2353840448 @default.
- W2510111515 hasRelatedWork W2406522397 @default.
- W2510111515 hasRelatedWork W2806866760 @default.
- W2510111515 hasRelatedWork W2970216048 @default.
- W2510111515 hasRelatedWork W4299822940 @default.
- W2510111515 isParatext "false" @default.
- W2510111515 isRetracted "false" @default.
- W2510111515 magId "2510111515" @default.
- W2510111515 workType "article" @default.