國立虎尾科技大學 |

Revealing and Analyzing the Shared Structure of Deep Face Embeddings.

Record Type:	Language materials, manuscript : Monograph/item
Title/Author:	Revealing and Analyzing the Shared Structure of Deep Face Embeddings./
Author:	McNeely-White, David G.
Description:	1 online resource (127 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 84-02, Section: B.
Contained By:	Dissertations Abstracts International84-02B.
Subject:	Computer science. -
Online resource:	click for full text (PQDT)
ISBN:	9798841783206

Revealing and Analyzing the Shared Structure of Deep Face Embeddings.
McNeely-White, David G.

Revealing and Analyzing the Shared Structure of Deep Face Embeddings. - 1 online resource (127 pages)

Source: Dissertations Abstracts International, Volume: 84-02, Section: B.

Thesis (Ph.D.)--Colorado State University, 2022.

Includes bibliographical references

Deep convolutional neural networks trained for face recognition are found to output face embeddings which share a fundamental structure. More specifically, one face verification model's embeddings (i.e. last--layer activations) can be compared directly to another model's embeddings after only a rotation or linear transformation, with little performance penalty. If only rotation is required to convert the bulk of embeddings between models, there is a strong sense in which those models are learning the same thing. In the most recent experiments, the structural similarity (and dissimilarity) of face embeddings is analyzed as a means of understanding face recognition bias. Bias has been identified in many face recognition models, often analyzed using distance measures between pairs of faces. By representing groups of faces as groups, and comparing them as groups, this shared embedding structure can be further understood. Specifically, demographic-specific subspaces are represented as points on a Grassmann manifold. Across 10 models, the geodesic distances between those points are expressive of demographic differences. By comparing how different groups of people are represented in the structure of embedding space, and how those structures vary with model designs, a new perspective on both representational similarity and face recognition bias is offered.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2024

Mode of access: World Wide Web

ISBN: 9798841783206Subjects--Topical Terms:

573171
Computer science.
Subjects--Index Terms:

Bias in face recognitionIndex Terms--Genre/Form:

554714
Electronic books.

Revealing and Analyzing the Shared Structure of Deep Face Embeddings.
LDR:02744ntm a22003977 4500 001 1147222
005 20240909100729.5
006 m o d
007 cr bn ---uuuuu
008 250605s2022 xx obm 000 0 eng d
020 $a 9798841783206
035 $a (MiAaPQ)AAI29209812
035 $a AAI29209812
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a McNeely-White, David G. $3 1472891
245 1 0 $a Revealing and Analyzing the Shared Structure of Deep Face Embeddings.
264 0 $c 2022
300 $a 1 online resource (127 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 84-02, Section: B.
500 $a Advisor: Beveridge, J. Ross.
502 $a Thesis (Ph.D.)--Colorado State University, 2022.
504 $a Includes bibliographical references
520 $a Deep convolutional neural networks trained for face recognition are found to output face embeddings which share a fundamental structure. More specifically, one face verification model's embeddings (i.e. last--layer activations) can be compared directly to another model's embeddings after only a rotation or linear transformation, with little performance penalty. If only rotation is required to convert the bulk of embeddings between models, there is a strong sense in which those models are learning the same thing. In the most recent experiments, the structural similarity (and dissimilarity) of face embeddings is analyzed as a means of understanding face recognition bias. Bias has been identified in many face recognition models, often analyzed using distance measures between pairs of faces. By representing groups of faces as groups, and comparing them as groups, this shared embedding structure can be further understood. Specifically, demographic-specific subspaces are represented as points on a Grassmann manifold. Across 10 models, the geodesic distances between those points are expressive of demographic differences. By comparing how different groups of people are represented in the structure of embedding space, and how those structures vary with model designs, a new perspective on both representational similarity and face recognition bias is offered.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2024
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
650 4 $a Social psychology. $3 554804
650 4 $a Social research. $3 1179269
650 4 $a Information science. $3 561178
653 $a Bias in face recognition
653 $a Face recognition
653 $a Grassmann
653 $a Representational similarity
655 7 $a Electronic books. $2 local $3 554714
690 $a 0984
690 $a 0344
690 $a 0723
690 $a 0451
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a Colorado State University. $b Computer Science. $3 1184420
773 0 $t Dissertations Abstracts International $g 84-02B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=29209812 $z click for full text (PQDT)