Claim Missing Document
Check
Articles

Found 2 Documents
Search
Journal : CommIT (Communication

Geometric Model for Human Body Orientation Classification Ardiyanto, Igi
CommIT (Communication and Information Technology) Journal Vol 9, No 1 (2015): CommIT Vol. 9 No. 1 Tahun 2015
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/commit.v9i1.1659

Abstract

This  paper proposes  an approach  for cal- culating  and estimating  human body orientation  using geometric model. A novel framework integrating gradient shape and texture model of the human body orientation is proposed.  The gradient  is a natural way for describing the human  shapes, while the texture  explains the body characteristic. The framework  is then combined with the random  forest classifier to obtain a robust  class  differ- ence  of the human body orientation. Experiments and comparison results are provided to show the advantages of our system over state-of-the-art. For both modeled and un-modeled gradient-texture  features with random forest classifier, they achieve the highest accuracy on separating each human orientation   class, respectively  56.9% and 67.3% for TUD-Stadtmitte  dataset.
Geometric Model for Human Body Orientation Classification Igi Ardiyanto
CommIT (Communication and Information Technology) Journal Vol. 9 No. 1 (2015): CommIT Journal
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/commit.v9i1.1659

Abstract

This  paper proposes  an approach  for cal- culating  and estimating  human body orientation  using geometric model. A novel framework integrating gradient shape and texture model of the human body orientation is proposed.  The gradient  is a natural way for describing the human  shapes, while the texture  explains the body characteristic. The framework  is then combined with the random  forest classifier to obtain a robust  class  differ- ence  of the human body orientation. Experiments and comparison results are provided to show the advantages of our system over state-of-the-art. For both modeled and un-modeled gradient-texture  features with random forest classifier, they achieve the highest accuracy on separating each human orientation   class, respectively  56.9% and 67.3% for TUD-Stadtmitte  dataset.