Recently, counting the number of people for crowd scenes is a hot topic because of its widespread applications (e.g. video surveillance, public security). It is a difficult task in the wild: changeable environment, large-range number of people cause the current methods can not work well. In addition, due to the scarce data, many methods suffer from over-fitting to a different extent. To remedy the above two problems, firstly, we develop a data collector and labeler, which can generate the synthetic crowd scenes and simultaneously annotate them without any manpower. Based on it, we build a large-scale, diverse synthetic dataset. Secondly, we propose two schemes that exploit the synthetic data to boost the performance of crowd counting in the wild: 1) pretrain a crowd counter on the synthetic data, then finetune it using the real data, which significantly prompts the model's performance on real data; 2) propose a crowd counting method via domain adaptation, which can free humans from heavy data annotations. Extensive experiments show that the first method achieves the state-of-the-art performance on four real datasets, and the second outperforms our baselines. The dataset and source code are available at

No abstract

Nano-composite Sm(0.5)Sr(0.5)CoO(3-δ) (SSC)-Ce(0.9)Gd(0.1)O(1.95) (GDC) and La(0.6)Sr(0.4)Co(0.8)Fe(0.2)O(3-δ) (LSCF)-GDC Solid Oxide Fuel Cell (SOFC) cathodes with various infiltrate loading levels were prepared through multiple nitrate solution infiltrations into porous GDC ionic conducting (IC) scaffolds. Microstructural analyses indicated that the average SSC and average LSCF hemispherical particle radii remained roughly constant, at 25 nm, across multiple infiltration-gelation-firing sequences. Comparisons between symmetric cell polarization resistance measurements and Simple Infiltrated Microstructure Polarization Loss Estimation (SIMPLE) model predictions showed that the SIMPLE model was able to predict the performance of heavily infiltrated SSC-GDC and LSCF-GDC cathodes with accuracies better than 55% and 70%, respectively (without the use of fitting parameters). Poor electronic conduction between mixed ionic electronic conducting (MIEC) infiltrate particles was found in lightly infiltrated cathodes. Since these electronic conduction losses were not accounted for by the SIMPLE model, larger discrepancies between the SIMPLE-model-predicted and measured polarization resistances were observed for lightly infiltrated cathodes. This work demonstrates that the SIMPLE model can be used to quickly determine the lowest possible polarization resistance of a variety of infiltrated MIEC on IC nano-composite cathodes (NCC's) when the NCC microstructure and an experimentally-applicable set of intrinsic MIEC oxygen surface resistances and IC bulk oxygen conductivities are known. Currently, this model is the only one capable of predicting the polarization resistance of heavily infiltrated MIEC on IC NCC's as a function of temperature, cathode thickness, nano-particle size, porosity, and composition.

No abstract

No abstract