Bounding overwatch

About: Bounding overwatch is a research topic. Over the lifetime, 966 publications have been published within this topic receiving 15156 citations.

Ellipsoidal Calculus for Estimation and Control

Abstract: This text gives an account of an ellipsoidal calculus and ellipsoidal techniques that allows presentation of the set-valued solutions to these problems in terms of approximating ellipsoidal-valued functions. Such an approach leads to effective computation schemes, an dopens the way to applications and implementations with computer animation, particularly in decision support systems. The problems treated here are those that involve calculation of attainability domains, of control synthesis under bounded controls, state constraints and unknown input disturbances, as well as those of "viability" and of the "bounding approach" to state estimation. The text ranges from a specially developed theory of exact set-valued solutions to the description of ellipsoidal calculus, related ellipsoidal-based methods and examples worked out with computer graphics. the calculus given here may also be interpreted as a generalized technique of the "interval analysis" type with an impact on scientific computation.

BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentation

Abstract: Recent leading approaches to semantic segmentation rely on deep convolutional networks trained with human-annotated, pixel-level segmentation masks. Such pixel-accurate supervision demands expensive labeling effort and limits the performance of deep networks that usually benefit from more training data. In this paper, we propose a method that achieves competitive accuracy but only requires easily obtained bounding box annotations. The basic idea is to iterate between automatically generating region proposals and training convolutional networks. These two steps gradually recover segmentation masks for improving the networks, and vise versa. Our method, called BoxSup, produces competitive results supervised by boxes only, on par with strong baselines fully supervised by masks under the same setting. By leveraging a large amount of bounding boxes, BoxSup further unleashes the power of deep convolutional networks and yields state-of-the-art results on PASCAL VOC 2012 and PASCAL-CONTEXT.

Multivariate Matching Methods That are Monotonic Imbalance Bounding

Abstract: We introduce a new ``Monotonic Imbalance Bounding'' (MIB) class of matching methods for causal inference that satisfies several important in-sample properties. MIB generalizes and extends in several new directions the only existing class, ``Equal Percent Bias Reducing'' (EPBR), which is designed to satisfy weaker properties and only in expectation. We also offer strategies to obtain specific members of the MIB class, and present a member of this class, called Coarsened Exact Matching, whose properties we analyze from this new perspective.

Efficient distance computation between non-convex objects

Abstract: This paper describes an efficient algorithm for computing the distance between nonconvex objects. Objects are modeled as the union of a set of convex components. From this model we construct a hierarchical bounding representation based on spheres. The distance between objects is determined by computing the distance between pairs of convex components using preexisting techniques. The key to efficiency is a simple search routine that uses the bounding representation to ignore most of the possible pairs of components. The efficiency can further be improved by accepting a relative error in the returned result. Several empirical trials are presented to examine the performance of the algorithm. >