An Implementation of Human Body Extraction Mechanism Based on Multi-Level Image Segmentation and Spline Regression From Single Images

Y. NUTHANA, A N NAGA JYOTHI

Abstract


Imaging of human body segments is demanding task which supports many applications such as understanding of scenes and recognition of activities. A bottom-up technology for extracting human bodies automatically from a single image, in case of almost upright position, is the available technique in cluttered environments. The dimension, position and face color are used for localizing human body, model construction of upper and lower body as per anthropometric constraints and skin color calculation. A highest level pose can be extracted by combining different levels of segmentation granularity. Jointly estimating the foreground and background during the body part search phase gives rise to the segments of the human body, that alleviates the need for shape matching exactly.A novel approach for extraction of standing human bodies has proposed in this paper where the highly dimensional pose space, scene density, and various human appearances are handled in a better way compared to conventional state of art methods. The proposed approach is classified into five different steps (a) face detection, (b) multi level segmentation, (c) skin detection, (d) upper body segmentation and (e) lower body segmentation respectively. Finally the simulation results have achieved better performance and high efficiency over traditional state of art methods.


Keywords


Multi level segmentation, skin detection, human bodies, super pixels, bottom-up approach

Full Text:

PDF

Refbacks

  • There are currently no refbacks.




Copyright (c) 2016 Edupedia Publications Pvt Ltd

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Publisher

EduPedia Publications Pvt Ltd, D/351, Prem Nanar-3, Suleman Nagar, Kirari, Nagloi, New Delhi PIN-Code 110086, India Through Phone Call us now: +919958037887 or +919557022047

All published Articles are Open Access at https://edupediapublications.org/journals/


Paper submission: editor@edupediapublications.com or edupediapublications@gmail.com

Editor-in-Chief       editor@edupediapublications.com

Mobile:                  +919557022047 & +919958037887

Websites   https://edupediapublications.org/journals/.

Journals Maintained and Hosted by

EduPedia Publications (P) Ltd in Association with Other Institutional Partners

http://edupediapublications.org/

Pen2Print and IJR are registered trademark of the Edupedia Publications Pvt Ltd.