Learning Streamed Attention Network from Descriptor Images for Cross-Resolution 3D Face Recognition

dc.contributor.authorNeto, João Baptista Cardia
dc.contributor.authorFerrari, Claudio
dc.contributor.authorMarana, Aparecido Nilceu [UNESP]
dc.contributor.authorBerretti, Stefano
dc.contributor.authorDel Bimbo, Alberto
dc.contributor.institutionSão Paulo State Technological College (FATEC)
dc.contributor.institutionUniversity of Parma
dc.contributor.institutionUniversidade Estadual Paulista (UNESP)
dc.contributor.institutionUniversity of Florence
dc.date.accessioned2023-07-29T16:05:41Z
dc.date.available2023-07-29T16:05:41Z
dc.date.issued2023-01-23
dc.description.abstractIn this article, we propose a hybrid framework for cross-resolution 3D face recognition which utilizes a Streamed Attention Network (SAN) that combines handcrafted features with Convolutional Neural Networks (CNNs). It consists of two main stages: first, we process the depth images to extract low-level surface descriptors and derive the corresponding Descriptor Images (DIs), represented as four-channel images. To build the DIs, we propose a variation of the 3D Local Binary Pattern (3DLBP) operator that encodes depth differences using a sigmoid function. Then, we design a CNN that learns from these DIs. The peculiarity of our solution consists in processing each channel of the input image separately, and fusing the contribution of each channel by means of both self- and cross-attention mechanisms. This strategy showed two main advantages over the direct application of Deep-CNN to depth images of the face; on the one hand, the DIs can reduce the diversity between high- and low-resolution data by encoding surface properties that are robust to resolution differences. On the other, it allows a better exploitation of the richer information provided by low-level features, resulting in improved recognition. We evaluated the proposed architecture in a challenging cross-dataset, cross-resolution scenario. To this aim, we first train the network on scanner-resolution 3D data. Next, we utilize the pre-trained network as feature extractor on low-resolution data, where the output of the last fully connected layer is used as face descriptor. Other than standard benchmarks, we also perform experiments on a newly collected dataset of paired high- and low-resolution 3D faces. We use the high-resolution data as gallery, while low-resolution faces are used as probe, allowing us to assess the real gap existing between these two types of data. Extensive experiments on low-resolution 3D face benchmarks show promising results with respect to state-of-the-art methods.en
dc.description.affiliationSão Paulo State Technological College (FATEC), Rua Maranhão, 898, Catanduva
dc.description.affiliationDepartment of Architecture and Engineering University of Parma, Parco Area delle Scienze, 181/A
dc.description.affiliationRecogna Laboratory São Paulo State University (UNESP), Av. Eng. Luís Edmundo Carrijo Coube
dc.description.affiliationMicc University of Florence, Viale Giovanni Battista Morgagni, 65
dc.description.affiliationUnespRecogna Laboratory São Paulo State University (UNESP), Av. Eng. Luís Edmundo Carrijo Coube
dc.identifierhttp://dx.doi.org/10.1145/3527158
dc.identifier.citationACM Transactions on Multimedia Computing, Communications and Applications, v. 19, n. 1, 2023.
dc.identifier.doi10.1145/3527158
dc.identifier.issn1551-6865
dc.identifier.issn1551-6857
dc.identifier.scopus2-s2.0-85148038627
dc.identifier.urihttp://hdl.handle.net/11449/249658
dc.language.isoeng
dc.relation.ispartofACM Transactions on Multimedia Computing, Communications and Applications
dc.sourceScopus
dc.subject3D face recognition
dc.subjectconvolutional neural networks
dc.subjectfeature descriptors
dc.subjectself- and cross-attention
dc.titleLearning Streamed Attention Network from Descriptor Images for Cross-Resolution 3D Face Recognitionen
dc.typeArtigo
unesp.campusUniversidade Estadual Paulista (Unesp), Faculdade de Ciências, Baurupt
unesp.departmentComputação - FCpt

Arquivos