No abstract

No abstract

No abstract

Traditional anthropometric evaluation needs professional measuring tools and operations, which is time-consuming, expensive, and not suitable for virtual try-on. As the mobile internet develops, the issue of human body reconstruction toward virtual try-on needs to be solved. This paper proposes a rapid human body reconstruction method for virtual try-on based on Multidimensional Dense Net (MDD-Net) on mobile terminal. MDD-Net takes the input of fusion features acquired by mobile as input and outputs 3D human body model to mobile supporting for virtual try-on. In the learning fuzzy anthropometric feature module, the example-guided fuzzy anthropometric feature matrix is acquired and default coding elements are interpolated. In the learning multi-perspective silhouette feature module, the fine human body shape features are learned based on DenseNet201. A corresponding fusion feature data set based on SMPL also is generated for MDD-Net training. In the experiments, without append fault-tolerant training samples, on the segmentation noise, nonstandard pose, and perspective error test set, the predicted accuracy of MDD-Net is improved by 13.34%, 55.77%, 34.6% and 43.4%, 37.2%, 9.0% respectively compared to Hs-Net and BfSNet proving its robust with the impact of uncertain positions and poses. And MDD-Net has a small error and standard deviation on critical anthropometric features explaining the effectiveness of our method.