No abstract

No abstract

No abstract

Convolutional neural networks(CNN) have been shown to perform better than the conventional stereo algorithms for stereo estimation. Numerous efforts focus on the pixel-wise matching cost computation, which is the important building block for many start-of-the-art algorithms. However, those architectures are limited to small and single scale receptive fields and use traditional methods for cost aggregation or even ignore cost aggregation. Differently we take them both into consideration. Firstly, we propose a new multi-scale matching cost computation sub-network, in which two different sizes of receptive fields are implemented parallelly. In this way, the network can make the best use of both variants and balance the trade-off between the increase of receptive field and the loss of detail. Furthermore, we show that our multi-dimension aggregation sub-network which containing 2D convolution and 3D convolution operations can provide rich context and semantic information for estimating an accurate initial disparity. Finally, experiments on challenging stereo benchmark KITTI demonstrate that the proposed method can achieve competitive results even without any additional post-processing.

The strength of levees can be affected during fluctuations in the water table. It is also possible for the climate to have an influence on the position of the water table in an earth levee. Traditional methods have resulted in approximate methods for dealing with the transient fluctuations of the water table in a levee. These approximations are generally accepted in engineering practice but the question can be rightfully raised as to how these approximations compare to a rigorous transient combined seepage and slope stability analysis. Software technology has significantly changed in recent years and is now at the point where it is much easier to perform transient seepage analyses. There are new questions that can be asked. Does an effective stress analysis diverge significantly from the 3-stage Duncan (1990) analysis? If so, under what conditions? This paper compares the Duncan (1990) three-stage methodology for analyzing rapid drawdown scenarios to a combined transient seepage and slope stability analysis. Traditional limit equilibrium methods will be utilized in the slope stability analysis and the accommodation of saturated and unsaturated pore-water pressures will be considered. Analyses of a number of typical cross-sections will be considered in order to determine the potential influence of geometry. The intent of the paper is to illustrate scenarios under which the Duncan (loc. cit.) methodology produces similar results to the results of a more rigorous analysis.