Objective measures for quality assessment of automatic skin enhancement algorithms

Mihai Ciuc; Adrian Capata; Corneliu Florea

doi:10.1117/12.838688

18 January 2010 Objective measures for quality assessment of automatic skin enhancement algorithms

Mihai Ciuc, Adrian Capata, Corneliu Florea

Proceedings Volume 7529, Image Quality and System Performance VII; 75290N (2010) https://doi.org/10.1117/12.838688
Event: IS&T/SPIE Electronic Imaging, 2010, San Jose, California, United States

Abstract

Automatic portrait enhancement by attenuating skin flaws (pimples, blemishes, wrinkles, etc.) has received considerable attention from digital camera manufacturers thanks to its impact on the public. Subsequently, a number of algorithms have been developed to meet this need. One central aspect to developing such an algorithm is quality assessment: having a few numbers that precisely indicate the amount of beautification brought by an algorithm (as perceived by human observers) is of great help, as it works on circumvent time-costly human evaluation. In this paper, we propose a method to numerically evaluate the quality of a skin beautification algorithm. The most important aspects we take into account and quantize to numbers are the quality of the skin detector, the amount of smoothing performed by the method, the preservation of intrinsic skin texture, and the preservation of facial features. We combine these measures into two numbers that assess the quality of skin detection and beautification. The derived measures are highly correlated with human perception, therefore they constitute a helpful tool for tuning and comparing algorithms.

Citation Download Citation

Mihai Ciuc, Adrian Capata, and Corneliu Florea "Objective measures for quality assessment of automatic skin enhancement algorithms", Proc. SPIE 7529, Image Quality and System Performance VII, 75290N (18 January 2010); https://doi.org/10.1117/12.838688

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available