scispace - formally typeset
Search or ask a question
Topic

Bounding overwatch

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


Papers
More filters
Book ChapterDOI
16 Jul 2007
TL;DR: A new multi-step kNN search algorithm that utilizes lower- and upper bounding distance information (Ilu) in the filter step is proposed and it is shown that, in contrast to existing approaches, the proposed solution is RIlu - optimal.
Abstract: Similarity search algorithms that directly rely on index structures and require a lot of distance computations are usually not applicable to databases containing complex objects and defining costly distance functions on spatial, temporal and multimedia data. Rather, the use of an adequate multi-step query processing strategy is crucial for the performance of a similarity search routine that deals with complex distance functions. Reducing the number of candidates returned from the filter step which then have to be exactly evaluated in the refinement step is fundamental for the efficiency of the query process. The state-of-the-art multi-step k-nearest neighbor (kNN) search algorithms are designed to use only a lower bounding distance estimation for candidate pruning. However, in many applications, also an upper bounding distance approximation is available that can additionally be used for reducing the number of candidates. In this paper, we generalize the traditional concept of R-optimality and introduce the notion of RI -optimality depending on the distance information I available in the filter step. We propose a new multi-step kNN search algorithm that utilizes lower- and upper bounding distance information (Ilu) in the filter step. Furthermore, we show that, in contrast to existing approaches, our proposed solution is RIlu - optimal. In an experimental evaluation, we demonstrate the significant performance gain over existing methods.

15 citations

Patent
28 Jun 2018
TL;DR: In this article, the authors used soft labeling to classify detection bounding boxes in images or phones of an input audio feature, where at least label has a range of possible values between 0 and 1.
Abstract: Apparatuses and methods of manufacturing same, systems, and methods for training deep learning machines are described. In one aspect, candidate units, such as detection bounding boxes in images or phones of an input audio feature, are classified using soft labelling, where at least label has a range of possible values between 0 and 1 based, in the case of images, on the overlap of a detection bounding box and one or more ground-truth bounding boxes for one or more classes.

15 citations

Book ChapterDOI
23 Apr 2008
TL;DR: This approach combines the compactness of OBBs and the simplicity of spheres by performing collision detection between static rigid objects using a bounding volume hierarchy which consists of an oriented bounding box (OBB) tree enhanced with bounding spheres.
Abstract: We perform collision detection between static rigid objects using a bounding volume hierarchy which consists of an oriented bounding box (OBB) tree enhanced with bounding spheres This approach combines the compactness of OBBs and the simplicity of spheres The majority of distant objects are separated using the simpler sphere tests The remaining objects are in close proximity, where some separation axes are significantly more effective than others We select 5 from among the 15 potential separating axes for OBBs Experimental results show that our algorithm achieved favorable speed up with respect to the existing OBB algorithms

15 citations

Journal ArticleDOI
28 May 2019
TL;DR: A method to combine bounding boxes extracted from multiple CNNs-based detections as a light and accurate alternative to confidence-based detection methods such as non-maximum suppression and clustering approaches predicting a single bounding box is introduced.
Abstract: In this paper, we present a multiple-object vehicle tracking system. We introduce a method to combine bounding boxes extracted from multiple CNNs-based detections as a light and accurate alternative to confidence-based detection methods such as non-maximum suppression and clustering approaches predicting a single bounding box. An intersection over union metric and a threshold value is proposed to determine whether a single detection or a connectivity graph between extracted bounding boxes exists. Affinity measurements are extracted from features representing bounding box geometry, appearance comparison, and changing scene properties. Then, data association is performed by solving the min-cost flow problem of the temporal windows’ affinity network. An affinity network of a directed graph associates the objects and determines whether an existing tracklet is maintained, terminated, or a new tracklet is initiated. Our model is evaluated and tested by KITTI object tracking—Car class benchmark dataset. Overall, the proposed multiple object tracking performance is ranked second according to the multiple object tracking accuracy, mostly tracked, mostly lost statistical metric values assure lower fragmentation and less than half of the captured ID-switch in comparison with respect to the method reaching at the highest multiple object tracking accuracy metric. Furthermore, the runtime is six times faster.

15 citations

Journal ArticleDOI
TL;DR: To the best of the knowledge, this is the first study about dual bounds ever derived for a commercial territory design problem and empirical evidence shows that the bounding scheme provides tighter lower bounds than those obtained by the linear programming relaxation.

15 citations


Network Information
Related Topics (5)
Robustness (computer science)
94.7K papers, 1.6M citations
85% related
Optimization problem
96.4K papers, 2.1M citations
85% related
Matrix (mathematics)
105.5K papers, 1.9M citations
82% related
Nonlinear system
208.1K papers, 4M citations
81% related
Artificial neural network
207K papers, 4.5M citations
80% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
2023714
20221,629
2021155
202075
201973
201850