1. Identity statement | |
Reference Type | Conference Paper (Conference Proceedings) |
Site | sibgrapi.sid.inpe.br |
Holder Code | ibi 8JMKD3MGPEW34M/46T9EHH |
Identifier | 8JMKD3MGPAW/3M5KU35 |
Repository | sid.inpe.br/sibgrapi/2016/07.23.00.14 |
Last Update | 2016:07.23.14.51.58 (UTC) administrator |
Metadata Repository | sid.inpe.br/sibgrapi/2016/07.23.00.14.14 |
Metadata Last Update | 2022:06.14.00.08.39 (UTC) administrator |
DOI | 10.1109/SIBGRAPI.2016.021 |
Citation Key | OliveiraMeSoJúPeGo:2016:DaAuMe |
Title | A Data Augmentation Methodology to Improve Age Estimation using Convolutional Neural Networks |
Format | On-line |
Year | 2016 |
Access Date | 2024, May 01 |
Number of Files | 1 |
Size | 1374 KiB |
|
2. Context | |
Author | 1 Oliveira, Ítalo de Pontes 2 Medeiros, João Lucas Peixoto 3 Sousa, Vinícius Fernandes de 4 Júnior, Adalberto Gomes Teixeira 5 Pereira, Eanes Torres 6 Gomes, Herman Martins |
Affiliation | 1 UFCG 2 UFCG 3 UFCG 4 UFCG 5 UFCG 6 UFCG |
Editor | Aliaga, Daniel G. Davis, Larry S. Farias, Ricardo C. Fernandes, Leandro A. F. Gibson, Stuart J. Giraldi, Gilson A. Gois, João Paulo Maciel, Anderson Menotti, David Miranda, Paulo A. V. Musse, Soraia Namikawa, Laercio Pamplona, Mauricio Papa, João Paulo Santos, Jefersson dos Schwartz, William Robson Thomaz, Carlos E. |
e-Mail Address | hmg@computacao.ufcg.edu.br |
Conference Name | Conference on Graphics, Patterns and Images, 29 (SIBGRAPI) |
Conference Location | São José dos Campos, SP, Brazil |
Date | 4-7 Oct. 2016 |
Publisher | IEEE Computer Society´s Conference Publishing Services |
Publisher City | Los Alamitos |
Book Title | Proceedings |
Tertiary Type | Full Paper |
History (UTC) | 2016-07-23 14:51:59 :: hmg@computacao.ufcg.edu.br -> administrator :: 2016 2016-10-05 14:49:19 :: administrator -> hmg@computacao.ufcg.edu.br :: 2016 2016-10-13 10:50:38 :: hmg@computacao.ufcg.edu.br -> administrator :: 2016 2022-06-14 00:08:39 :: administrator -> :: 2016 |
|
3. Content and structure | |
Is the master or a copy? | is the master |
Content Stage | completed |
Transferable | 1 |
Version Type | finaldraft |
Keywords | data augmentation age estimation deep learning fiducial points face detection |
Abstract | Recent advances in deep learning methodologies are enabling the construction of more accurate classifiers. However, existing labeled face datasets are limited in size, which prevents CNN models from reaching their full generalization capabilities. A variety of techniques to generate new training samples based on data augmentation have been proposed, but the great majority is limited to very simple transformations. The approach proposed in this paper takes into account intrinsic information about human faces in order to generate an augmented dataset that is used to train a CNN, by creating photo-realistic smooth face variations based on Active Appearance Models optimized for human faces. An experimental evaluation taking CNN models trained with original and augmented versions of the MORPH face dataset allowed an increase of 10% in the F-Score and yielded Receiver Operating Characteristic curves that outperformed state-of-the-art work in the literature. |
Arrangement 1 | urlib.net > SDLA > Fonds > SIBGRAPI 2016 > A Data Augmentation... |
Arrangement 2 | urlib.net > SDLA > Fonds > Full Index > A Data Augmentation... |
doc Directory Content | access |
source Directory Content | PID4373767.pdf | 22/07/2016 21:14 | 1.3 MiB | |
agreement Directory Content | |
|
4. Conditions of access and use | |
data URL | http://urlib.net/ibi/8JMKD3MGPAW/3M5KU35 |
zipped data URL | http://urlib.net/zip/8JMKD3MGPAW/3M5KU35 |
Language | en |
Target File | PID4374341.pdf |
User Group | hmg@computacao.ufcg.edu.br |
Visibility | shown |
Update Permission | not transferred |
|
5. Allied materials | |
Mirror Repository | sid.inpe.br/banon/2001/03.30.15.38.24 |
Next Higher Units | 8JMKD3MGPAW/3M2D4LP 8JMKD3MGPEW34M/4742MCS |
Citing Item List | sid.inpe.br/sibgrapi/2016/07.02.23.50 5 |
Host Collection | sid.inpe.br/banon/2001/03.30.15.38 |
|
6. Notes | |
Empty Fields | archivingpolicy archivist area callnumber contenttype copyholder copyright creatorhistory descriptionlevel dissemination edition electronicmailaddress group isbn issn label lineage mark nextedition notes numberofvolumes orcid organization pages parameterlist parentrepositories previousedition previouslowerunit progress project readergroup readpermission resumeid rightsholder schedulinginformation secondarydate secondarykey secondarymark secondarytype serieseditor session shorttitle sponsor subject tertiarymark type url volume |
|