Convolutional neural networks are capable of learning powerful representational spaces, which are necessary for tackling complex learning tasks. However, due to the model capacity required to capture such representations, they are often susceptible to overfitting and therefore require proper regularization in order to generalize well. In this paper, we show that the simple regularization techni...

A drop impacting a target cutout in a thin polymer film is wrapped by the film in a dynamic sequence involving both capillary forces and inertia. Different 3D structures can be produced from a given target by slightly varying the impact parameters. A simplified model for a nonlinear dynamic Elastica coupled with a drop successfully explains this shape selection and yields detailed quantitative ...

We present a technique for steganography in polygonal meshes. Our method hides a message in the indexed representation of a mesh by permuting the order in which faces and vertices are stored. The permutation is relative to a reference ordering that encoder and decoder derive from the mesh connectivity in a consistent manner. Our method is distortion-free because it does not modify the geometry ...

We address the problem of building 3D object models from image sequences obtained with known camera motion. An approach based on a local reconstruction method is presented. Recovered surfaces are described as polygonal meshes. To this purpose, reconstructed points are triangulated and surface areas which are not covered by rims are detected since they may lead to false reconstructed points. Res...

The vibratory bowl feeder is the oldest and still most common approach to the automated feeding (orienting) of industrial parts. In this paper we consider a class of vibratory bowl filters that can be described by removing polygonal sections from the track; we refer to this class of filters as traps. For an n-sided polygonal part and an m-sided polygonal trap, we give an O(nm log n) algorithm t...

We develop a valuation theory for generalized polygons similar to the existing theory for dense near polygons. This valuation theory has applications for the study and classification of generalized polygons that have full subpolygons as subgeometries.

Earlier [I], it was shown that certain reinforced holes may be designed for elastic, isotropic plane sheets that do not alter the stress state in the cut sheet, for particular types of planar loading. In this work, this methodology is extended to symmetrically laminated composite plates which are under planar loading, and additionally pure bending moments. Since these two types of loading lead ...

To evaluate a new technique of Dynamic hip screw (DHS) fixation without using traction table. From july 2006 to july 2012, 328 patients undergoing DHS fixation using manual traction technique were evaluated in terms of Tip apex distance (TAD), -quality of fracture reduction, operation time, preoperative preparation time. In 97% cases good to moderate reduction could be achieved. The mean TAD wa...

This paper revisits the existence and construction problems for polygonal designs (a special class of partially balanced incomplete block designs associated with regular polygons). We present new polygonal designs with various parameter sets by explicit construction. In doing so we employ several construction methods – some conventional and some new. We also establish a link between a class of ...

A near-polygonal graph is a graph Γ which has a set C of m-cycles for some positive integer m such that each 2-path of Γ is contained in exactly one cycle in C. If m is the girth of Γ , then the graph is called polygonal. We provide a construction of an infinite family of polygonal graphs of arbitrary even girth with 2-arc transitive automorphism groups, showing that there are infinitely many 2...