FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4304080720> ?p ?o ?g. }

• W4304080720 abstract "Cross-resolution face recognition (CRFR) in an open-set setting is a practical application for surveillance scenarios where low-resolution (LR) probe faces captured via surveillance cameras require being matched to a watchlist of high-resolution (HR) galleries. Although CRFR is to be of practical use, it sees a performance drop of more than 10% compared to that of high-resolution face recognition protocols. The challenges of CRFR are multifold, including the domain gap induced by the HR and LR images, the pose/texture variations, etc. To this end, this work systematically discusses possible issues and their solutions that affect the accuracy of CRFR. First, we explore the effect of resolution changes and conclude that resolution matching is the key for CRFR. Even simply downscaling the HR faces to match the LR ones brings a performance gain. Next, to further boost the accuracy of matching cross-resolution faces, we found that a well-designed super-resolution network, which can (a) represent the images continuously, is (b) suitable for real-world degradation kernel, (c) adaptive to different input resolutions, and (d) guided by an identity-preserved loss, is necessary to upsample the LR faces with discriminative enhancement. Here, the proposed identity-preserved loss plays the role of reconciling the objective discrepancy of super-resolution between human perception and machine recognition. Finally, we emphasize that removing the pose variations is an essential step before matching faces for recognition in the super-resolved feature space. Our method is evaluated on benchmark datasets, including SCface, cross-resolution LFW, and QMUL-Tinyface. The results show that the proposed method outperforms the SOTA methods by a clear margin and narrows the performance gap compared to the high-resolution face recognition protocol." @default.
• W4304080720 created "2022-10-10" @default.
• W4304080720 creator A5024171815 @default.
• W4304080720 creator A5040050806 @default.
• W4304080720 creator A5059613626 @default.
• W4304080720 creator A5074744839 @default.
• W4304080720 creator A5084360712 @default.
• W4304080720 date "2022-10-10" @default.
• W4304080720 modified "2023-10-16" @default.
• W4304080720 title "Towards Understanding Cross Resolution Feature Matching for Surveillance Face Recognition" @default.
• W4304080720 cites W2770121394 @default.
• W4304080720 cites W2792481260 @default.
• W4304080720 cites W2798877493 @default.
• W4304080720 cites W2938076880 @default.
• W4304080720 cites W2962898354 @default.
• W4304080720 cites W2963566548 @default.
• W4304080720 cites W2963839617 @default.
• W4304080720 cites W2964167901 @default.
• W4304080720 cites W2969985801 @default.
• W4304080720 cites W2970747919 @default.
• W4304080720 cites W2971225816 @default.
• W4304080720 cites W2997694845 @default.
• W4304080720 cites W3018976951 @default.
• W4304080720 cites W3024539255 @default.
• W4304080720 cites W3034302825 @default.
• W4304080720 cites W3038701588 @default.
• W4304080720 cites W3040291497 @default.
• W4304080720 cites W3090533328 @default.
• W4304080720 cites W3093367965 @default.
• W4304080720 cites W3101766778 @default.
• W4304080720 cites W3101998545 @default.
• W4304080720 cites W3119712657 @default.
• W4304080720 cites W3124399606 @default.
• W4304080720 cites W3160336571 @default.
• W4304080720 cites W3167584510 @default.
• W4304080720 cites W3169129566 @default.
• W4304080720 cites W3174865552 @default.
• W4304080720 cites W3184311248 @default.
• W4304080720 cites W3187746867 @default.
• W4304080720 cites W3203631022 @default.
• W4304080720 cites W3205301887 @default.
• W4304080720 cites W3205609346 @default.
• W4304080720 cites W4210496811 @default.
• W4304080720 doi "https://doi.org/10.1145/3503161.3548402" @default.
• W4304080720 hasPublicationYear "2022" @default.
• W4304080720 type Work @default.
• W4304080720 citedByCount "0" @default.
• W4304080720 crossrefType "proceedings-article" @default.
• W4304080720 hasAuthorship W4304080720A5024171815 @default.
• W4304080720 hasAuthorship W4304080720A5040050806 @default.
• W4304080720 hasAuthorship W4304080720A5059613626 @default.
• W4304080720 hasAuthorship W4304080720A5074744839 @default.
• W4304080720 hasAuthorship W4304080720A5084360712 @default.
• W4304080720 hasConcept C105795698 @default.
• W4304080720 hasConcept C13280743 @default.
• W4304080720 hasConcept C138268822 @default.
• W4304080720 hasConcept C138885662 @default.
• W4304080720 hasConcept C144024400 @default.
• W4304080720 hasConcept C153180895 @default.
• W4304080720 hasConcept C154945302 @default.
• W4304080720 hasConcept C165064840 @default.
• W4304080720 hasConcept C185798385 @default.
• W4304080720 hasConcept C205649164 @default.
• W4304080720 hasConcept C2776401178 @default.
• W4304080720 hasConcept C2779304628 @default.
• W4304080720 hasConcept C31510193 @default.
• W4304080720 hasConcept C31972630 @default.
• W4304080720 hasConcept C33923547 @default.
• W4304080720 hasConcept C36289849 @default.
• W4304080720 hasConcept C41008148 @default.
• W4304080720 hasConcept C41895202 @default.
• W4304080720 hasConcept C97931131 @default.
• W4304080720 hasConceptScore W4304080720C105795698 @default.
• W4304080720 hasConceptScore W4304080720C13280743 @default.
• W4304080720 hasConceptScore W4304080720C138268822 @default.
• W4304080720 hasConceptScore W4304080720C138885662 @default.
• W4304080720 hasConceptScore W4304080720C144024400 @default.
• W4304080720 hasConceptScore W4304080720C153180895 @default.
• W4304080720 hasConceptScore W4304080720C154945302 @default.
• W4304080720 hasConceptScore W4304080720C165064840 @default.
• W4304080720 hasConceptScore W4304080720C185798385 @default.
• W4304080720 hasConceptScore W4304080720C205649164 @default.
• W4304080720 hasConceptScore W4304080720C2776401178 @default.
• W4304080720 hasConceptScore W4304080720C2779304628 @default.
• W4304080720 hasConceptScore W4304080720C31510193 @default.
• W4304080720 hasConceptScore W4304080720C31972630 @default.
• W4304080720 hasConceptScore W4304080720C33923547 @default.
• W4304080720 hasConceptScore W4304080720C36289849 @default.
• W4304080720 hasConceptScore W4304080720C41008148 @default.
• W4304080720 hasConceptScore W4304080720C41895202 @default.
• W4304080720 hasConceptScore W4304080720C97931131 @default.
• W4304080720 hasLocation W43040807201 @default.
• W4304080720 hasOpenAccess W4304080720 @default.
• W4304080720 hasPrimaryLocation W43040807201 @default.
• W4304080720 hasRelatedWork W1652783584 @default.
• W4304080720 hasRelatedWork W1844323538 @default.
• W4304080720 hasRelatedWork W1982770690 @default.
• W4304080720 hasRelatedWork W2136485282 @default.
• W4304080720 hasRelatedWork W2353840448 @default.
• W4304080720 hasRelatedWork W2404514746 @default.

• next