Next Article in Journal
On Charge Conjugation, Chirality and Helicity of the Dirac and Majorana Equation for Massive Leptons
Next Article in Special Issue
Reduction by Lie Group Symmetries in Diffeomorphic Image Registration and Deformation Modelling
Previous Article in Journal
Confluent Supersymmetric Partners of Quantum Systems Emerging from the Spheroidal Equation
Previous Article in Special Issue
The Perception of Symmetry in Depth: Effect of Symmetry Plane Orientation
Open AccessArticle

Unsupervised Object Modeling and Segmentation with Symmetry Detection for Human Activity Recognition

1
Department of Computer Science and Engineering, National Taiwan Ocean University, 2 Pei-Ning Road, Keelung 202, Taiwan
2
Department of New Media and Communications Administration, Ming Chuan University, 250 Sec. 5 Zhong Shan North Road, Taipei 111, Taiwan
*
Author to whom correspondence should be addressed.
Academic Editor: Christopher Tyler
Symmetry 2015, 7(2), 427-449; https://doi.org/10.3390/sym7020427
Received: 29 November 2014 / Revised: 17 March 2015 / Accepted: 16 April 2015 / Published: 23 April 2015
(This article belongs to the Special Issue Symmetry: Theory and Applications in Vision)
In this paper we present a novel unsupervised approach to detecting and segmenting objects as well as their constituent symmetric parts in an image. Traditional unsupervised image segmentation is limited by two obvious deficiencies: the object detection accuracy degrades with the misaligned boundaries between the segmented regions and the target, and pre-learned models are required to group regions into meaningful objects. To tackle these difficulties, the proposed approach aims at incorporating the pair-wise detection of symmetric patches to achieve the goal of segmenting images into symmetric parts. The skeletons of these symmetric parts then provide estimates of the bounding boxes to locate the target objects. Finally, for each detected object, the graphcut-based segmentation algorithm is applied to find its contour. The proposed approach has significant advantages: no a priori object models are used, and multiple objects are detected. To verify the effectiveness of the approach based on the cues that a face part contains an oval shape and skin colors, human objects are extracted from among the detected objects. The detected human objects and their parts are finally tracked across video frames to capture the object part movements for learning the human activity models from video clips. Experimental results show that the proposed method gives good performance on publicly available datasets. View Full-Text
Keywords: object detection and segmentation; Hough voting; human activity recognition; symmetry detection object detection and segmentation; Hough voting; human activity recognition; symmetry detection
MDPI and ACS Style

Su, J.-Y.; Cheng, S.-C.; Huang, D.-K. Unsupervised Object Modeling and Segmentation with Symmetry Detection for Human Activity Recognition. Symmetry 2015, 7, 427-449.

Show more citation formats Show less citations formats

Article Access Map by Country/Region

1
Only visits after 24 November 2015 are recorded.
Back to TopTop