Graduate School and Research Center in Digital Sciences

The impact of privacy protection filters on gender recognition

Ruchaud, Natacha; Antipov, Grigory; Korshunov, Pavel; Dugelay, Jean-Luc; Ebrahimi, Touradj; Berrani, Sid-Ahmed

OPTICS+PHOTONICS 2015, SPIE Optical Engineering + Applications, Applications of Digital Image Processing XXXVIII, August 10-13, 2015, San Diego, CA, USA

Deep learning-based algorithms have become increasingly ecient in recognition and detection tasks, especially when they are trained on large-scale datasets. Such recent success has led to a speculation that deep learning methods are comparable to or even outperform human visual system in its ability to detect and recognize objects and their features. In this paper, we focus on the speci c task of gender recognition in images when they have been processed by privacy protection lters (e.g., blurring, masking, and pixelization) applied at di erent strengths. Assuming a privacy protection scenario, we compare the performance of state of the art deep learning algorithms with a subjective evaluation obtained via crowdsourcing to understand how privacy protection lters a ect both machine and human vision.

Document Doi Hal Bibtex

Title:The impact of privacy protection filters on gender recognition
Keywords:Gender Recognition, Deep Learning, Crowdsourcing
Type:Conference
Language:English
City:San Diego
Country:UNITED STATES
Date:
Department:Digital Security
Eurecom ref:4667
Copyright: © 2015 Society of Photo-Optical Instrumentation Engineers. This paper is published in OPTICS+PHOTONICS 2015, SPIE Optical Engineering + Applications, Applications of Digital Image Processing XXXVIII, August 10-13, 2015, San Diego, CA, USA and is made available as an electronic preprint with permission of SPIE. One print or electronic copy may be made for personal use only. Systematic or multiple reproduction, distribution to multiple locations via electronic or other means, duplication of any material in this paper for a fee or for commercial purposes, or modification of the content of the paper are prohibited.
Bibtex: @inproceedings{EURECOM+4667, doi = {http://dx.doi.org/10.1117/12.2193647}, year = {2015}, title = {{T}he impact of privacy protection filters on gender recognition}, author = {{R}uchaud, {N}atacha and {A}ntipov, {G}rigory and {K}orshunov, {P}avel and {D}ugelay, {J}ean-{L}uc and {E}brahimi, {T}ouradj and {B}errani, {S}id-{A}hmed}, booktitle = {{OPTICS}+{PHOTONICS} 2015, {SPIE} {O}ptical {E}ngineering + {A}pplications, {A}pplications of {D}igital {I}mage {P}rocessing {XXXVIII}, {A}ugust 10-13, 2015, {S}an {D}iego, {CA}, {USA}}, address = {{S}an {D}iego, {UNITED} {STATES}}, month = {08}, url = {http://www.eurecom.fr/publication/4667} }
See also: