Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385804902> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W4385804902 abstract "Modern algorithms for RGB-IR facial recognition— a challenging problem where infrared probe images are matched with visible gallery images—leverage precise and accurate guidance from curated (i.e., labeled) data to bridge large spectral differences. However, supervised cross-spectral face recognition methods are often extremely sensitive due to over-fitting to labels, performing well in some settings but not in others. Moreover, when fine-tuning on data from additional settings, supervised cross-spectral face recognition are prone to catastrophic forgetting. Therefore, we propose a novel unsupervised framework for RGB-IR face recognition to minimize the cost and time inefficiencies pertaining to labeling large-scale, multispectral data required to train supervised cross-spectral recognition methods and to alleviate the effect of forgetting by removing over dependence on hard labels to bridge such large spectral differences. The proposed framework integrates an efficient backbone network architecture with part-based attention models, which collectively enhances common information between visible and infrared faces. Then, the framework is optimized using pseudo-labels and a new cross-spectral memory bank loss. This framework is evaluated on the ARL-VTF and TUFTS datasets, achieving 98.55% and 43.28% true accept rate, respectively. Additionally, we analyze effects of forgetting and show that our framework is less prone to these effects." @default.
- W4385804902 created "2023-08-15" @default.
- W4385804902 creator A5075289214 @default.
- W4385804902 creator A5079769754 @default.
- W4385804902 creator A5083865668 @default.
- W4385804902 creator A5084163563 @default.
- W4385804902 date "2023-06-01" @default.
- W4385804902 modified "2023-10-14" @default.
- W4385804902 title "Mitigating Catastrophic Interference using Unsupervised Multi-Part Attention for RGB-IR Face Recognition" @default.
- W4385804902 cites W1971359334 @default.
- W4385804902 cites W1991830825 @default.
- W4385804902 cites W2104068492 @default.
- W4385804902 cites W2341788950 @default.
- W4385804902 cites W2405210883 @default.
- W4385804902 cites W2485822847 @default.
- W4385804902 cites W2485847574 @default.
- W4385804902 cites W2902038035 @default.
- W4385804902 cites W2904427185 @default.
- W4385804902 cites W2904856314 @default.
- W4385804902 cites W2963073614 @default.
- W4385804902 cites W2963156682 @default.
- W4385804902 cites W2963609056 @default.
- W4385804902 cites W2963639219 @default.
- W4385804902 cites W3006477128 @default.
- W4385804902 cites W3101862707 @default.
- W4385804902 cites W3102910076 @default.
- W4385804902 cites W3118642508 @default.
- W4385804902 cites W3119532581 @default.
- W4385804902 cites W3120285682 @default.
- W4385804902 cites W3132834324 @default.
- W4385804902 cites W3141689296 @default.
- W4385804902 doi "https://doi.org/10.1109/cvprw59228.2023.00039" @default.
- W4385804902 hasPublicationYear "2023" @default.
- W4385804902 type Work @default.
- W4385804902 citedByCount "0" @default.
- W4385804902 crossrefType "proceedings-article" @default.
- W4385804902 hasAuthorship W4385804902A5075289214 @default.
- W4385804902 hasAuthorship W4385804902A5079769754 @default.
- W4385804902 hasAuthorship W4385804902A5083865668 @default.
- W4385804902 hasAuthorship W4385804902A5084163563 @default.
- W4385804902 hasConcept C138885662 @default.
- W4385804902 hasConcept C144024400 @default.
- W4385804902 hasConcept C153083717 @default.
- W4385804902 hasConcept C153180895 @default.
- W4385804902 hasConcept C154945302 @default.
- W4385804902 hasConcept C173163844 @default.
- W4385804902 hasConcept C2779304628 @default.
- W4385804902 hasConcept C31510193 @default.
- W4385804902 hasConcept C31972630 @default.
- W4385804902 hasConcept C36289849 @default.
- W4385804902 hasConcept C41008148 @default.
- W4385804902 hasConcept C41895202 @default.
- W4385804902 hasConcept C7149132 @default.
- W4385804902 hasConcept C82990744 @default.
- W4385804902 hasConceptScore W4385804902C138885662 @default.
- W4385804902 hasConceptScore W4385804902C144024400 @default.
- W4385804902 hasConceptScore W4385804902C153083717 @default.
- W4385804902 hasConceptScore W4385804902C153180895 @default.
- W4385804902 hasConceptScore W4385804902C154945302 @default.
- W4385804902 hasConceptScore W4385804902C173163844 @default.
- W4385804902 hasConceptScore W4385804902C2779304628 @default.
- W4385804902 hasConceptScore W4385804902C31510193 @default.
- W4385804902 hasConceptScore W4385804902C31972630 @default.
- W4385804902 hasConceptScore W4385804902C36289849 @default.
- W4385804902 hasConceptScore W4385804902C41008148 @default.
- W4385804902 hasConceptScore W4385804902C41895202 @default.
- W4385804902 hasConceptScore W4385804902C7149132 @default.
- W4385804902 hasConceptScore W4385804902C82990744 @default.
- W4385804902 hasFunder F4320332179 @default.
- W4385804902 hasFunder F4320333051 @default.
- W4385804902 hasLocation W43858049021 @default.
- W4385804902 hasOpenAccess W4385804902 @default.
- W4385804902 hasPrimaryLocation W43858049021 @default.
- W4385804902 hasRelatedWork W1775397219 @default.
- W4385804902 hasRelatedWork W2283162247 @default.
- W4385804902 hasRelatedWork W2347601237 @default.
- W4385804902 hasRelatedWork W2524507886 @default.
- W4385804902 hasRelatedWork W2897995864 @default.
- W4385804902 hasRelatedWork W2908959303 @default.
- W4385804902 hasRelatedWork W2956043259 @default.
- W4385804902 hasRelatedWork W4210401281 @default.
- W4385804902 hasRelatedWork W4212983513 @default.
- W4385804902 hasRelatedWork W4303182286 @default.
- W4385804902 isParatext "false" @default.
- W4385804902 isRetracted "false" @default.
- W4385804902 workType "article" @default.