Automatically detecting protruding objects when shooting environmental portraits

Authors:
Pei-Yu Lo;Sheng-Wen Shih;Jen-Chang Liu;Jen-Shin Hong
Affiliations:
Department of Computer Science and Information Engineering, National Chi Nan University, Nantou, Taiwan;Department of Computer Science and Information Engineering, National Chi Nan University, Nantou, Taiwan;Department of Computer Science and Information Engineering, National Chi Nan University, Nantou, Taiwan;Department of Computer Science and Information Engineering, National Chi Nan University, Nantou, Taiwan
Venue:
ACCV'10 Proceedings of the 2010 international conference on Computer vision - Volume part II
Year:
2010

Citing 5
Cited 0

A Computational Approach to Edge Detection

IEEE Transactions on Pattern Analysis and Machine Intelligence
Watersheds in Digital Spaces: An Efficient Algorithm Based on Immersion Simulations

IEEE Transactions on Pattern Analysis and Machine Intelligence
Say cheese! Experiences with a robot photographer

AI Magazine
Towards intelligent photo composition-automatic detection of unintentional dissection lines in environmental portrait photos

Expert Systems with Applications: An International Journal
Computational models and experimental investigations of effects of balance and symmetry on the aesthetics of text-overlaid images

International Journal of Human-Computer Studies

Quantified Score

Hi-index	0.00

Visualization

Abstract

This study proposes techniques for detecting unintentional protruding objects from a subject's head in environmental portraits. The protruding objects are determined based on the color and edge information of the background regions adjacent to the head regions in an image sequence. The proposed algorithm consists of watershed segmentation and KLT feature tracking model for extracting foreground regions, a ROI (Region of Interest) extracting model based on face detection results, and a protruding object detection model based on the color clusters and edges of the background regions inside the ROI. Experimental evaluations using four test videos with different backgrounds, lighting conditions, and head ornaments show that the average detection rate and false detection rate of the proposed algorithm are 87.40% and 12.11% respectively.