With the techonologic improvements in neural networks, researchers have developed a methodology to create 3D models of human from 2D images. The project is titled as: PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization.
In artificial network researches, one of the main topics is 3D imaging or image processing. Of course this is determined by the customer requests. Specifically in this field, there are a lot of groups that are working on 3D human shape reconstruction. Although current approaches have demonstrated the potential in real world settings, they still fail to produce reconstructions with the level of detail often present in the input images.
The limiting factor for 3D human image reconstruction is the size of the image. As the number of pixels increases, the processing time increases a lot. Due to this, teams tend to work with pictures with lower resolution. As a result of this, the reconstruction quality degrades.
We address this limitation by formulating a multi-level architecture that is end-to-end trainable. A coarse level observes the whole image at lower resolution and focuses on holistic reasoning. This provides context to an fine level which estimates highly detailed geometry by observing higher-resolution images. We demonstrate that our approach significantly outperforms existing state-of-the-art techniques on single image human shape reconstruction by fully leveraging 1k-resolution input images.