izpis_h1_title_alt

Deep Face Decoder : towards understanding the embedding space of convolutional networks through visual reconstruction of deep face templates
ID Križaj, Janez (Avtor), ID Plesh, Richard O. (Avtor), ID Banavar, Mahesh (Avtor), ID Schuckers, Stephanie (Avtor), ID Štruc, Vitomir (Avtor)

.pdfPDF - Predstavitvena datoteka, prenos (6,68 MB)
MD5: 06824F9A044039547890B396A6E81DDE
URLURL - Izvorni URL, za dostop obiščite https://www.sciencedirect.com/science/article/pii/S095219762400099X Povezava se odpre v novem oknu

Izvleček
Advances in deep learning and convolutional neural networks (ConvNets) have driven remarkable face recognition (FR) progress recently. However, the black-box nature of modern ConvNet-based face recognition models makes it challenging to interpret their decision-making process, to understand the reasoning behind specific success and failure cases, or to predict their responses to unseen data characteristics. It is, therefore, critical to design mechanisms that explain the inner workings of contemporary FR models and offer insight into their behavior. To address this challenge, we present in this paper a novel template-inversion approach capable of reconstructing high-fidelity face images from the embeddings (templates, feature-space representations) produced by modern FR techniques. Our approach is based on a novel Deep Face Decoder (DFD) trained in a regression setting to visualize the information encoded in the embedding space with the goal of fostering explainability. We utilize the developed DFD model in comprehensive experiments on multiple unconstrained face datasets, namely Visual Geometry Group Face dataset 2 (VGGFace2), Labeled Faces in the Wild (LFW), and Celebrity Faces Attributes Dataset High Quality (CelebA-HQ). Our analysis focuses on the embedding spaces of two distinct face recognition models with backbones based on the Visual Geometry Group 16-layer model (VGG-16) and the 50-layer Residual Network (ResNet-50). The results reveal how information is encoded in the two considered models and how perturbations in image appearance due to rotations, translations, scaling, occlusion, or adversarial attacks, are propagated into the embedding space. Our study offers researchers a deeper comprehension of the underlying mechanisms of ConvNet-based FR models, ultimately promoting advancements in model design and explainability.

Jezik:Angleški jezik
Ključne besede:deep face templates, template inversion, face recognition, deep learning, explainability, interpretability
Vrsta gradiva:Članek v reviji
Tipologija:1.01 - Izvirni znanstveni članek
Organizacija:FE - Fakulteta za elektrotehniko
Status publikacije:Objavljeno
Različica publikacije:Objavljena publikacija
Leto izida:2024
Št. strani:20 str.
Številčenje:Vol. 132, art. 107941
PID:20.500.12556/RUL-156219 Povezava se odpre v novem oknu
UDK:004.93
ISSN pri članku:1873-6769
DOI:10.1016/j.engappai.2024.107941 Povezava se odpre v novem oknu
COBISS.SI-ID:184095491 Povezava se odpre v novem oknu
Datum objave v RUL:14.05.2024
Število ogledov:78
Število prenosov:29
Metapodatki:XML RDF-CHPDL DC-XML DC-RDF
:
Kopiraj citat
Objavi na:Bookmark and Share

Gradivo je del revije

Naslov:Engineering applications of artificial intelligence
Založnik:Elsevier
ISSN:1873-6769
COBISS.SI-ID:23000325 Povezava se odpre v novem oknu

Licence

Licenca:CC BY 4.0, Creative Commons Priznanje avtorstva 4.0 Mednarodna
Povezava:http://creativecommons.org/licenses/by/4.0/deed.sl
Opis:To je standardna licenca Creative Commons, ki daje uporabnikom največ možnosti za nadaljnjo uporabo dela, pri čemer morajo navesti avtorja.

Sekundarni jezik

Jezik:Slovenski jezik
Ključne besede:globoke obrazne predloge, inverzija predlog, prepoznavanje obrazov, globoko učenje, razložljivost, interpretabilnost

Projekti

Financer:ARRS - Agencija za raziskovalno dejavnost Republike Slovenije
Številka projekta:P2-0250
Naslov:Metrologija in biometrični sistemi

Financer:ARRS - Agencija za raziskovalno dejavnost Republike Slovenije
Številka projekta:J2-2501
Naslov:Globoki generativni modeli za lepotno in modno industrijo (DeepBeauty)

Financer:Drugi - Drug financer ali več financerjev
Program financ.:Center for Identification Technology Research

Financer:NSF - National Science Foundation
Številka projekta:1650503

Podobna dela

Podobna dela v RUL:
Podobna dela v drugih slovenskih zbirkah:

Nazaj