ABSTRACT
Point clouds acquired by 3D scanning devices are often sparse, noisy, and non-uniform, causing a loss of geometric features. To facilitate the usability of point clouds in downstream applications, given such input, we present a learning-based point upsampling method, i.e., which generates dense and uniform points at arbitrary ratios and better captures sharp features. To generate feature-aware points, we introduce cross fields that are aligned to sharp geometric features by self-supervision to guide point generation. Given cross field defined frames, we enable arbitrary ratio upsampling by learning at each input point a local parameterized surface. The learned surface consumes the neighboring points and 2D tangent plane coordinates as input, and maps onto a continuous surface in 3D where arbitrary ratios of output points can be sampled. To solve the non-uniformity of input points, on top of the cross field guided upsampling, we further introduce an iterative strategy that refines the point distribution by moving sparse points onto the desired continuous 3D surface in each iteration. Within only a few iterations, the sparse points are evenly distributed and their corresponding dense samples are more uniform and better capture geometric features. Through extensive evaluations on diverse scans of objects and scenes, we demonstrate that iPUNet is robust to handle noisy and non-uniformly distributed inputs, and outperforms state-of-the-art point cloud upsampling methods.
ABSTRACT
Oxygen evolution reaction (OER) electrocatalysts in acidic media, except for precious IrO2, have difficulty realizing good electrocatalytic activity and high electrochemical stability simultaneously. However, the scarcity of IrO2 as an acidic OER electrocatalyst impedes its large-scale application in hydrogen generation, organic synthesis, nonferrous metal production and sewage disposal. Herein, we report the design and fabrication of a nanoporous TiMnCoCN compound based on the nanoscale Kirkendall effect, possessing an intriguing self-adjusting capability for the oxygen evolution reaction (OER) in a 0.5 M H2SO4 solution. The nanoporous TiMnCoCN compound electrode for the acidic OER displays a low overpotential of 143 mV for 10 mA cm-2 and exhibits no increase in potential over 50,000 s, which is ascribed to the self-adjusting ability, Carbon/nitrogen (C/N) incorporation and nanoporous architecture. The concentration of inert TiO2 on the reconstructed surface of the compound can self-adjust with the change in OER potential via a cobalt-dissolved vacancy approach according to the stabilization requirement. In this work, the self-reconstruction law of surface structure was discovered, providing a novel strategy for designing and fabricating nonnoble OER electrocatalysts with superior catalytic performance and robust stability in acidic media.
ABSTRACT
Ammonia (NH3) emission is one of the major environmental issues in livestock farming. Gas measurements are required to study the emission process, to establish emission factors, and to assess the efficiency of emission reduction techniques. However, the current methods for acquiring reference measurements of NH3 are either high in cost or labor intensive. In this study, a cost-effective ammonia monitoring system (AMS) was constructed from a commercially-available gas analyzing module based on tunable diode laser absorption (TDLA) spectroscopy. To cope with the negative measurement biases caused by differing inlet pressures, a set of correction equations was formulated. Field validation of the AMS on NH3 measurement was conducted in a fattening pig barn, where the system was compared to a Fourier-transform infrared (FTIR) spectroscopy analyzer. Under two test conditions in a fattening pig barn, the absolute error of the AMS measurements with respect to the average obtained values between the AMS and the FTIR was respectively 0.66 and 0.08 ppmv, corresponding to 5.9% and 0.5% relative error. Potential sources of the measurement uncertainties in both the AMS and FTIR were discussed. The test results demonstrated that the AMS was capable of performing high-quality measurement with sub-ppm accuracy, making it a promising cost-effective tool for establishing NH3 emission factors and studying NH3 emission processes in pig houses.
