This model can definitely help AR developers.
Are you specialized in AR projects? You might like this new research, focused on recreating hands from RGB images, that was originally discussed in the paper "Towards Accurate Alignment in Real-time 3D Hand-Mesh Reconstruction" and then presented at ICCV 2021.
3D hand-mesh reconstruction from RGB images can be used in different cases, including augmented reality (AR) and mixed reality experiences, but the task requires real-time speed, accurate hand pose, plus plausible mesh-image alignment, and meeting all three usually turns out to be a challenge for developers.
"With careful design in the network structure and in the loss functions, we can promote high-quality finger-level mesh-image alignment and drive the models together to deliver real-time predictions," states the abstract. "Extensive quantitative and qualitative results on benchmark datasets demonstrate that the quality of our results outperforms the state-of-the-art methods on hand-mesh/pose precision and hand-image alignment. In the end, we also showcase several real-time AR scenarios."