Face detection is a fundamental problem in computer vision. It is still a challenging task in unconstrained conditions due to significant variations in scale, pose, expressions, and occlusion. In this paper, we propose a multi-branch fully convolutional network (MB-FCN) for face detection, which considers both efficiency and effectiveness in the design process. Our MB-FCN detector can deal with...

We consider the problem of designing hierarchical two layer ring networks. The top layer consists of a federal-ring which establishes connection between a number of node disjoint metro-rings in a bottom layer. The objective is to minimize the costs of links in the network, taking both the fixed link establishment costs and the link capacity costs into account. Hierarchical ring network design p...

We consider efficient Bayesian network structure learning (BNSL) based on scores using branch and bound. Thus far, as a BNSL score, the Bayesian Dirichlet equivalent uniform (BDeu) has been used most often, but it is recently proved that the BDeu does not choose the simplest model even when the likelihood is maximized whereas Jeffreys’ prior and MDL satisfy such regularity. Although the BDeu ha...

This paper addresses the segmentation from an image of entities that have the form of a ‘network’, i.e. the region in the image corresponding to the entity is composed of branches joining together at junctions, e.g. road or vascular networks. We present a new phase field higher-order active contour (HOAC) prior model for network regions, and apply it to the segmentation of road networks from ve...

We present a branch-and-cut algorithm to solve capacitated network design problems. Given a capacitated network and point-to-point traac demands, the objective is to install more capacity on the edges of the network and route traac simultaneously, so that the overall cost is minimized. We study a mixed-integer programming formulation of the problem and identify some new facet deening inequaliti...

Nowadays, most classification networks use one-hot encoding to represent categorical data because of its simplicity. However, may affect the generalization ability as it neglects inter-class correlations. We observe that, even when a neural network trained with labels produces incorrect predictions, still pays attention target image region and reveals which classes confuse network. Inspired by ...

In this paper, I examine the role of banks in propagating local economic shocks from one area to another through their network of bank branches, by exploiting a newly developed branch-level dataset. Specifically, I examine the change in the geographical distribution of small business loans within each bank network in response to: 1) increases in deposit growth due to presence in areas with new ...

Here we offer a more detailed description of the proposed model to facilitate re-implementation. A schematic illustration of the network architecture of the proposed model can be found in Fig. 2 of the main paper. It shows that the model is a Siamese triplet network with three branches of identical architectures and shared parameters. In this section, we further describe the detailed network ar...

Axial migration of red blood cells in small glass tubes can cause blood viscosity to be effectively independent of shear rate. However, this phase separation may not occur to the same degree in the venous network due to infusion of cells and aggregates at branch points. To investigate this hypothesis, we followed trajectories of fluorescently labeled red blood cells in the venular network of th...

A credal network associates sets of probability distributions with directed acyclic graphs. Under strong independence assumptions, inference with credal networks is equivalent to a signomial program under linear constraints, a problem that is NP-hard even for categorical variables and polytree models. We describe an approach for inference with polytrees that is based on branch-and-bound optimiz...