Comparing Principal and Independent Modes of Variation in 3D Human Torso Shape Using PCA and ICA

نویسندگان

  • Anthony Ruto
  • Mike Lee
  • Bernard Buxton
چکیده

We analyse 3D human torso data using Principal Components Analysis (PCA) and Independent Components Analysis (ICA) and compare their respective principal and independent modes of variation. Both PCA and ICA have been used to analyse variations in observed data for different applications. PCA offers a means of capturing the significant variations present in a data sample while ICA is useful in identifying “blind sources” or contributing factors that produce the observed variation. We use both methods to analyse a set of female 3D human torso data and study how the two sets of modes describe the shape, size and other variations in the data. Thirty PCA modes describing 90% of the total variation in the sample were used, together with the corresponding 30 ICA modes calculated therefrom using FastICA. Sample mode weight coefficients from both sets are evaluated and correlated with age, weight and a set of twelve key torso measurements to determine how individual PCA and ICA modes correlate with body measurements such as those used in the clothing industry and in human anthropometry. When the two sets of modes are visualised, the individual ICA modes are seen to identify more specific variations in the overall size, shape and posture variation in the sample than the individual PCA modes. When compared with age, weight and the key torso measurements, the measurements were found to have strong correlations with a small number of the most significant PCA modes whilst the correlations with each of the ICA modes were weaker, spread over a much larger number of modes, and could be used to group the measurements into clusters suggestive of different types of torso variation. This is consistent with the fact that each ICA mode encapsulates a significant range of shape variation whilst only the first few, most significant PCA modes do so. We conclude therefore that ICA offers a means of identifying specific variations in 3D human torso shape that better describe how torso shape varies. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. c ©2006 The University of Liverpool

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

An accurate 3D human face model reconstruction scheme

The purpose of this thesis was to develop a 3-dimensional (3D) reconstruction algorithm which can more accurately generate a 3D shape model of a human face from a single 2-dimensional (2D) image. Like most present day approaches to building 3D human facial models, the proposed algorithm stemmed from the statistical approach pioneered by Blanz and Vetter [13], who used a morphable 3D model which...

متن کامل

Active Shape Model of Combining Pca and Ica: Application to Facial Feature Extraction

Active Shape Model (ASM) is a powerful statistical tool to extract the facial features of a face image under frontal view. It mainly relies on Principle Component Analysis (PCA) to statistically model the variability in the training set of example shapes. Independent Component Analysis (ICA) has been proven to be more efficient to extract face features than PCA . In this paper, we combine the P...

متن کامل

Human Figure Segmentation Using Independent Component Analysis

In this paper, we present a Statistical Shape Model for Human Figure Segmentation in gait sequences. Point Distribution Models (PDM) generally use Principal Component analysis (PCA) to describe the main directions of variation in the training set. However, PCA assumes a number of restrictions on the data that do not always hold. In this work, we explore the potential of Independent Component An...

متن کامل

Independent Component Analysis in Statistical Shape Models

Statistical shape models generally use Principal Component Analysis (PCA) to describe the main directions of shape variation in a training set of example shapes. However, PCA assumes a number of restrictions on the data that do not always hold. In this paper we explore the use of an alternative shape decomposition, Independent Component Analysis (ICA), which does not assume a Gaussian distribut...

متن کامل

ICA vs. PCA Active Appearance Models: Application to Cardiac MR Segmentation

Statistical shape models generally use Principal Component Analysis (PCA) to describe the main directions of shape variation in a training set of example shapes. However, PCA has the restriction that the input data must be drawn from a Gaussian distribution and is only able to describe global shape variations. In this paper we evaluate the use of an alternative shape decomposition, Independent ...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2006