We propose a straightforward method that simultaneously reconstructs the 3D facial structure and provides dense alignment. To achieve this, we design a 2D representation called UV position map which records the 3D shape of a complete face in UV space, then train a simple Convolutional Neural Network to regress it from a single 2D image. We also integrate a weight mask into the loss function during training to improve the performance of the network. Our method does not rely on any prior face model, and can reconstruct full facial geometry along with semantic meaning. Meanwhile, our network is very light-weighted and spends only 9.8ms to process an image, which is extremely faster than previous works. Experiments on multiple challenging datasets show that our method surpasses other state-of-the-art methods on both reconstruction and alignment tasks by a large margin. Code is available at https://github.com/YadiraF/PRNet.

No abstract

Gold nanomaterials have attracted considerable interest as vehicles for intracellular drug delivery. In our study, we synthesized three different shapes of methylpolyethylene glycol coated-anisotropic gold nanoparticles: stars, rods, and triangles. The cellular internalization of these nanoparticles by RAW264.7 cells was analyzed, providing a parametric evaluation of the effect of shape. The efficiency of cellular uptake of the gold nanoparticles was found to rank in the following order from lowest to highest: stars, rods, and triangles. The possible mechanisms of cellular uptake for the three types of gold nanoparticles were examined, and it was found that different shapes tended to use the various endocytosis pathways in different proportions. Our study, which has demonstrated that shape can modulate the uptake of nanoparticles into RAW264.7 cells and that triangles were the shape with the most efficient cellular uptake, provides useful guidance toward the design of nanomaterials for drug delivery.

We have shown from both simulations and experiments that zwitterion functionalized carbon nanotubes (CNTs) can be used to construct highly efficient desalination membranes. Our simulations predicted that zwitterion functional groups at the ends of CNTs allow a high flux of water, while rejecting essentially all ions. We have synthesized zwitterion functionalized CNT/polyamide nanocomposite membranes with varying loadings of CNTs and assessed these membranes for water desalination. The CNTs within the polyamide layer were partially aligned through a high-vacuum filtration step during membrane synthesis. Addition of zwitterion functionalized CNTs into a polyamide membrane increased both the flux of water and the salt rejection ratio. The flux of water was found to increase by more than a factor of 4, from 6.8 to 28.7 GFD (gallons per square foot per day), as the fraction of CNTs was increased from 0 to 20 wt %. Importantly, the ion rejection ratio increased slightly from 97.6% to 98.6%. Thus, the nanotubes imparted an additional transport mechanism to the polyamide membrane, having higher flow rate and the same or slightly better selectivity. Simulations show that when two zwitterions are attached to each end of CNTs having diameters of about 15 Å, the ion rejection ratio is essentially 100%. In contrast, the rejection ratio for nonfunctionalized CNTs is about 0%, and roughly 20% for CNTs having five carboxylic acid groups per end. The increase in ion rejection for the zwitterion functionalized CNTs is due to a combination of steric hindrance from the functional groups partially blocking the tube ends and electrostatic repulsion between functional groups and ions, with steric effects dominating. Theoretical predictions indicate that an ideal CNT/polymer membrane having a loading of 20 wt % CNTs would have a maximum flux of about 20000 GFD at the conditions of our experiments.

No abstract