Research tasks related to human body analysis have been drawing a lot of attention in computer vision area over the last few decades, considering its potential benefits on our day-to-day life. Anthropometry is a field defining physical measures of a human body size, form, and functional capacities. Specifically, the accurate estimation of anthropometric body measurements from visual human body data is one of the challenging problems, where the solution would ease many different areas of applications, including ergonomics, garment manufacturing, etc. This paper formulates a research in the field of deep learning and neural networks, to tackle the challenge of body measurements estimation from various types of visual input data (such as 2D images or 3D point clouds). Also, we deal with the lack of real human data annotated with ground truth body measurements required for training and evaluation, by generating a synthetic dataset of various human body shapes and performing a skeleton-driven annotation.
Figure 1: (From left to right) Our method takes an input model and computes a skeleton using an iterative Laplacian smoothing. Then, using an association between each skeleton segment and vertices around it the method maps each skeleton segment into rectangular textures. Finally, the method computes a global surface to texture mapping by packing the rectangular textures into a final texture.
AbstractIn the article an idea for a novel way of mapping of textures onto a surface of 3D model is introduced. Our technique is based on two interlocking mappings. The first one maps surface vertices onto a computed skeleton and the second one maps the surrounding area of each skeleton segment into a rectangle with size based on the surface properties around the segment. Furthermore, these rectangles are packed into a squared texture called skeleton texture map (STM) by approximately solving a palette loading problem. Our technique enables the mapping of a texture onto the surface without necessity to store texture coordinates with the model data and it is also suitable for surfaces with a topology non-homotopic to a sphere with higher order genus and unlimited structure branching.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.