Single sample face recognition from video via stacked supervised auto-encoder

Pedro J.Soto Vega; Raul Queiroz Feitosa; Victor H.Ayma Quirita; Patrick Nigri Happ

doi:10.1109/SIBGRAPI.2016.022

Single sample face recognition from video via stacked supervised auto-encoder

Pedro J.Soto Vega
, Raul Queiroz Feitosa
, Victor H.Ayma Quirita
, Patrick Nigri Happ

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

12 Scopus citations

Abstract

This work proposes and evaluates strategies based on Stacked Supervised Auto-Encoders (SSAE) for face representation in video surveillance applications. The study focuses on the identification task with a single sample per person (SSPP) in the gallery. Variations in terms of pose, facial expression, illumination and occlusion are approached in two ways. First, the SSAE extracts features from face images, which are robust to such variations. Second, we propose methods to exploit the multiple samples per persons probes (MSPPP) that can be extracted from video sequences. Three variants of the proposed method are compared upon HONDA/UCSD and VIDTIMIT video datasets. The experimental results demonstrate that strategies combining SSAE and MSPPP are able to outperform other SSPP methods, such a local binary patterns, in face recognition from video.

Original language	English
Title of host publication	2016 29th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)
Place of Publication	United Estates
Pages	96-103
Number of pages	8
ISBN (Electronic)	9781509035687
DOIs	https://doi.org/10.1109/SIBGRAPI.2016.022
State	Published - 16 Jan 2017
Externally published	Yes
Event	SIBGRAPI Conference on Graphics, Patterns and Images - Sao Paulo, Brazil Duration: 4 Oct 2016 → 7 Oct 2016 Conference number: 29 http://gibis.unifesp.br/sibgrapi16/

Publication series

Name	Brazilian Symposium of Computer Graphic and Image Processing
ISSN (Print)	1530-1834
ISSN (Electronic)	2377-5416

Conference

Conference	SIBGRAPI Conference on Graphics, Patterns and Images
Abbreviated title	SIBGRAPI
Country/Territory	Brazil
City	Sao Paulo
Period	4/10/16 → 7/10/16
Internet address	http://gibis.unifesp.br/sibgrapi16/

Bibliographical note

Publisher Copyright: © 2016 IEEE.

SIBGRAPI - Conference on Graphics, Patterns and Images is an international conference annually promoted by the Brazilian Computer Society (SBC).

Date Added to IEEE Xplore: 16 January 2017

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 9 Industry, Innovation, and Infrastructure
SDG 17 Partnerships for the Goals

Keywords

Auto-encoder
Face Recognition
Surveillance

Access to Document

10.1109/SIBGRAPI.2016.022

Cite this

@inproceedings{b16a6c38f0ab4ed984598e0045a7df3d,

title = "Single sample face recognition from video via stacked supervised auto-encoder",

abstract = "This work proposes and evaluates strategies based on Stacked Supervised Auto-Encoders (SSAE) for face representation in video surveillance applications. The study focuses on the identification task with a single sample per person (SSPP) in the gallery. Variations in terms of pose, facial expression, illumination and occlusion are approached in two ways. First, the SSAE extracts features from face images, which are robust to such variations. Second, we propose methods to exploit the multiple samples per persons probes (MSPPP) that can be extracted from video sequences. Three variants of the proposed method are compared upon HONDA/UCSD and VIDTIMIT video datasets. The experimental results demonstrate that strategies combining SSAE and MSPPP are able to outperform other SSPP methods, such a local binary patterns, in face recognition from video.",

keywords = "Auto-encoder, Face Recognition, Surveillance",

author = "Vega, \{Pedro J.Soto\} and Feitosa, \{Raul Queiroz\} and Quirita, \{Victor H.Ayma\} and Happ, \{Patrick Nigri\}",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; SIBGRAPI Conference on Graphics, Patterns and Images ; Conference date: 04-10-2016 Through 07-10-2016",

year = "2017",

month = jan,

day = "16",

doi = "10.1109/SIBGRAPI.2016.022",

language = "Ingl{\'e}s",

isbn = "9781509035694",

series = "Brazilian Symposium of Computer Graphic and Image Processing",

pages = "96--103",

booktitle = "2016 29th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)",

url = "http://gibis.unifesp.br/sibgrapi16/",

}

Vega, PJS, Feitosa, RQ, Quirita, VHA & Happ, PN 2017, Single sample face recognition from video via stacked supervised auto-encoder. in 2016 29th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)., 7813021, Brazilian Symposium of Computer Graphic and Image Processing, United Estates, pp. 96-103, SIBGRAPI Conference on Graphics, Patterns and Images , Sao Paulo, Brazil, 4/10/16. https://doi.org/10.1109/SIBGRAPI.2016.022

Single sample face recognition from video via stacked supervised auto-encoder. / Vega, Pedro J.Soto; Feitosa, Raul Queiroz; Quirita, Victor H.Ayma et al.
2016 29th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI). United Estates, 2017. p. 96-103 7813021 (Brazilian Symposium of Computer Graphic and Image Processing).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Single sample face recognition from video via stacked supervised auto-encoder

AU - Vega, Pedro J.Soto

AU - Feitosa, Raul Queiroz

AU - Quirita, Victor H.Ayma

AU - Happ, Patrick Nigri

N1 - Conference code: 29

PY - 2017/1/16

Y1 - 2017/1/16

N2 - This work proposes and evaluates strategies based on Stacked Supervised Auto-Encoders (SSAE) for face representation in video surveillance applications. The study focuses on the identification task with a single sample per person (SSPP) in the gallery. Variations in terms of pose, facial expression, illumination and occlusion are approached in two ways. First, the SSAE extracts features from face images, which are robust to such variations. Second, we propose methods to exploit the multiple samples per persons probes (MSPPP) that can be extracted from video sequences. Three variants of the proposed method are compared upon HONDA/UCSD and VIDTIMIT video datasets. The experimental results demonstrate that strategies combining SSAE and MSPPP are able to outperform other SSPP methods, such a local binary patterns, in face recognition from video.

AB - This work proposes and evaluates strategies based on Stacked Supervised Auto-Encoders (SSAE) for face representation in video surveillance applications. The study focuses on the identification task with a single sample per person (SSPP) in the gallery. Variations in terms of pose, facial expression, illumination and occlusion are approached in two ways. First, the SSAE extracts features from face images, which are robust to such variations. Second, we propose methods to exploit the multiple samples per persons probes (MSPPP) that can be extracted from video sequences. Three variants of the proposed method are compared upon HONDA/UCSD and VIDTIMIT video datasets. The experimental results demonstrate that strategies combining SSAE and MSPPP are able to outperform other SSPP methods, such a local binary patterns, in face recognition from video.

KW - Auto-encoder

KW - Face Recognition

KW - Surveillance

UR - https://www.scopus.com/pages/publications/85013746162

U2 - 10.1109/SIBGRAPI.2016.022

DO - 10.1109/SIBGRAPI.2016.022

M3 - Contribución a la conferencia

AN - SCOPUS:85013746162

SN - 9781509035694

T3 - Brazilian Symposium of Computer Graphic and Image Processing

SP - 96

EP - 103

BT - 2016 29th SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI)

CY - United Estates

T2 - SIBGRAPI Conference on Graphics, Patterns and Images

Y2 - 4 October 2016 through 7 October 2016

ER -

Single sample face recognition from video via stacked supervised auto-encoder

Abstract

Publication series

Conference

Bibliographical note

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this