Please use this identifier to cite or link to this item: https://hdl.handle.net/10356/78020
Title: Special effect processing of portraits
Authors: Ling, Xi Wen
Keywords: DRNTU::Engineering::Electrical and electronic engineering
Issue Date: 2019
Abstract: With the proliferation of smartphones with cameras, more and more digital photos are taken every year. People may take photos of themselves to record a moment, a selfie to post on social media, or many other purposes. However, most people are casual photographers who do not know how to use photo editing software to retouch their photos and cover up skin imperfections. Therefore, an interactive app that automatically implements special effect to artistically modify a portrait photo could be useful for them. To retouch facial skin remotely from the background, the first step was to apply skin segmentation algorithm on the portrait photo to identify facial skin. To improve the accuracy of correct identified skin pixels, morphological operations and additional masks were used. The next step was to apply Gaussian filter to the identified skin pixels to blur skin imperfections on the face. Gradual boundary filtering was applied in this step to the face outlines to output a more natural looking portrait. The final filtered photo improved the human subject’s facial skin condition without affecting the facial features or background. This recreated the effect of a professionally retouched portrait photo. More special effects could be implemented in the future using the same process flow.
URI: http://hdl.handle.net/10356/78020
Schools: School of Electrical and Electronic Engineering 
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
FYP Final Report.pdf
  Restricted Access
3.1 MBAdobe PDFView/Open

Page view(s)

259
Updated on Jun 22, 2024

Download(s)

13
Updated on Jun 22, 2024

Google ScholarTM

Check

Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.