A limited comparative study of dimension reduction techniques on CAESAR

Patrick, James A.; Clouse, Hamilton Scott; Mendoza-Schrock, Olga; Arnold, Gregory

doi:10.1109/naecon.2010.5712939

Cited by 2 publications

(1 citation statement)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To this end, they examined MRI data from healthy and diseased human muscle tissue and attempted to classify the samples accordingly [4]. Patrick et al (2010) explored dimensionality reduction techniques over the Civillian American and European Surface Anthropometry Resource Project (CEASAR) database [11]. In this work, their objective was to compare the methods of diffusion map and principle components analysis (PCA) with the k-nearest neighbor and Adaboost classifiers for gender classification.…”

Section: Related Workmentioning

confidence: 99%

Diffusion maps for exploring electro-optical synthetic vehicle image data

Ramirez

Mendoza-Schrock

2012

2012 IEEE National Aerospace and Electronics Conference (NAECON)

Self Cite

View full text Add to dashboard Cite

In this work, we explore low-dimensional representations of high-dimensional data derived from electro-optical synthetic vehicle images. The collection of vehicle images consists of four different vehicle models: Toyota Camry, Toyota Avalon, Toyota Tacoma, and Nissan Sentra. This data contains 3,601 160 × 213 gray-scale vehicle images sampled uniformly over a camera view hemisphere. We use the non-linear manifold learning technique of diffusion maps with Gaussian kernel to explore low-dimensional structure the high-dimensional cloud of vehicle image observations. Diffusion maps have been shown to be a valuable tool in the analysis of high-dimensional data and the technique is able to extract an approximation for the underlying structure inherent to the data. Our analysis includes examining how the diffusion time and kernel width leads to different lowdimensional representations and we present a novel technique to relate the kernel width to the distribution of data in the observation space. In addition, we present initial results for multiclass vehicle classification through low-dimensional embedding coordinates and the out-of-sample extension of unlabeled vehicle images.

show abstract

Section: Related Workmentioning

confidence: 99%

Diffusion maps for exploring electro-optical synthetic vehicle image data

Ramirez

Mendoza-Schrock

2012

2012 IEEE National Aerospace and Electronics Conference (NAECON)

Self Cite

View full text Add to dashboard Cite

show abstract

Novel Anthropometry Based on 3D-Bodyscans Applied to a Large Population Based Cohort

et al. 2016

View full text Add to dashboard Cite

Three-dimensional (3D) whole body scanners are increasingly used as precise measuring tools for the rapid quantification of anthropometric measures in epidemiological studies. We analyzed 3D whole body scanning data of nearly 10,000 participants of a cohort collected from the adult population of Leipzig, one of the largest cities in Eastern Germany. We present a novel approach for the systematic analysis of this data which aims at identifying distinguishable clusters of body shapes called body types. In the first step, our method aggregates body measures provided by the scanner into meta-measures, each representing one relevant dimension of the body shape. In a next step, we stratified the cohort into body types and assessed their stability and dependence on the size of the underlying cohort. Using self-organizing maps (SOM) we identified thirteen robust meta-measures and fifteen body types comprising between 1 and 18 percent of the total cohort size. Thirteen of them are virtually gender specific (six for women and seven for men) and thus reflect most abundant body shapes of women and men. Two body types include both women and men, and describe androgynous body shapes that lack typical gender specific features. The body types disentangle a large variability of body shapes enabling distinctions which go beyond the traditional indices such as body mass index, the waist-to-height ratio, the waist-to-hip ratio and the mortality-hazard ABSI-index. In a next step, we will link the identified body types with disease predispositions to study how size and shape of the human body impact health and disease.

show abstract

A limited comparative study of dimension reduction techniques on CAESAR

Cited by 2 publications

References 11 publications

Diffusion maps for exploring electro-optical synthetic vehicle image data

Diffusion maps for exploring electro-optical synthetic vehicle image data

Novel Anthropometry Based on 3D-Bodyscans Applied to a Large Population Based Cohort

Contact Info

Product

Resources

About