Abstract
Human body pose and shape estimation is an important and challenging task in computer vision. This paper presents a novel method for estimating 3D human body pose and shape from several RGB images, using detected joint positions in the images and based on a parametric human body model. Firstly, the 2D joint points of the RGB images are estimated using a deep neural network, which provides a strong prior on the pose. Then, an energy function is constructed based on the 2D joint points in the RGB images and a parametric human body model. By minimizing the energy function, the pose, shape and camera parameters are obtained. The main contribution of the method over previous work, is that the optimization is based on several images simultaneously using only estimated joint positions in the images. We have performed experiments on both synthetic and real image data-sets, that demonstrate that our method can reconstruct 3D human bodies with better accuracy than previous single view methods.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have