Detailed 3D human body reconstruction from multi-view images combining voxel super-resolution and learned implicit representation

Abstract

The task of reconstructing detailed 3D human body models from images is interesting but challenging in computer vision due to the high freedom of human bodies. This work proposes a coarse-to-fine method to reconstruct detailed 3D human body from multi-view images combining Voxel Super-Resolution (VSR) based on learning the implicit representation. Firstly, the coarse 3D models are estimated by learning an Pixel-aligned Implicit Function based on Multi-scale Features (MF-PIFu) which are extracted by multi-stage hourglass networks from the multi-view images. Then, taking the low resolution voxel grids which are generated by the coarse 3D models as input, the VSR is implemented by learning an implicit function through a multi-stage 3D convolutional neural network. Finally, the refined detailed 3D human body models can be produced by VSR which can preserve the details and reduce the false reconstruction of the coarse 3D models. Benefiting from the implicit representation, the training process in our method is memory efficient and the detailed 3D human body produced by our method from multi-view images is the continuous decision boundary with high-resolution geometry. In addition, the coarse-to-fine method based on MF-PIFu and VSR can remove false reconstructions and preserve the appearance details in the final reconstruction, simultaneously. In the experiments, our method quantitatively and qualitatively achieves the competitive 3D human body models from images with various poses and shapes on both the real and synthetic datasets.

Emergence of 3D body model-based infant health evaluation: a review

“…Benefits and Limitations of using 3D body models 3D body models offer a highly detailed and accurate portrayal of an infant's body, facilitating the assessment of health and identification of potential health issues more precisely [54][55][56] . They enable monitoring of growth and development over time, aiding early detection of health concerns 57,58 .…”

A single-image human body reconstruction method combining normal recovery and frequency domain completion

“…With the rapid evolution of virtual reality, human-computer interaction, and digital entertainment, realistic 3D human body models have emerged as a pivotal resource for creating immersive experiences in various applications. Compared to the 3D human body reconstruction method based on multiple images [1][2][3][4] , although a single image [5][6][7] provides limited depth information, it is more practical in application due to its ease of acquisition and its ability to avoid inconsistencies and matching problems between multiple images. Considering the high degree of freedom in human body movements and the wide variety of clothing, reconstructing a 3D human body from a single image remains an extremely challenging task.…”