Everyday spatio-temporal reasoning is driven through qualitative abstractions over mental maps or 'diagrams'. Diagrammatic reasoning involves direct manipulation and inspection of diagrams as the primary means...
详细信息
ISBN:
(纸本)9781467385640
Everyday spatio-temporal reasoning is driven through qualitative abstractions over mental maps or 'diagrams'. Diagrammatic reasoning involves direct manipulation and inspection of diagrams as the primary means of inference. Diagrammatic representation offer computational advantage in problems where spatial relationships play a prominent role. In video, objects change spatial relationships over time. therefore, combining diagrammatic reasoning with qualitative spatial and temporal reasoning holds promise. In this paper, we put forward a framework combining diagrammatic representation with qualitative spatial and temporal reasoning for motion event detection in video. Key frames with tracked objects for a given video are extracted. these frames in forward moving time are represented using specified;'diagrams'. A set of perception function is defined exploiting results from diagrammatic reasoning and qualitative spatial and temporal reasoning for determining spatial relations between objects of interest. Inter diagrammatic reasoning operator is used to combine sequence of diagrams for extracting spatio-temporal changes. Diagram modification function is defined exploiting results from qualitative reasoning to extract directional information. Considering extracted relative position and relative direction of displacement as features, we use supervised machine learning techniques to recognize motion events in video. the approach is tested in videos with few people/groups meeting, walking together and splitting up/fighting from the CAVIAR dataset.
Over 70% of software development effort is spent in software maintenance comprising bug fixes and version updates. these activities involve fast comprehension of large codebases authored by multiple developers. Develo...
详细信息
this paper presents a real-time hands-free immersive image navigation system that can respond to various gestures and voice commands. We combine Microsoft Kinect 2.0 and Leap Motion Controller, on a single platform, t...
详细信息
the proceedings contain 15 papers. the topics discussed include: development of the logic programming approach to the intelligent monitoring of anomalous human behaviour;location of pupil contour by Hough transform of...
ISBN:
(纸本)9789897580949
the proceedings contain 15 papers. the topics discussed include: development of the logic programming approach to the intelligent monitoring of anomalous human behaviour;location of pupil contour by Hough transform of connectivity components;testing an image mining approach to obtain pressure ulcers stage and texture;on image representing in image analysis;a variational method to remove the combination of Poisson and Gaussian noises;PRIAR using a graph segmentation method;virtual immersive environments for underwater archaeological exploration;current trends in mathematical image analysis - a survey;human pose estimation in video via MCMC sampling;signal processing for underwater archaeology;experimenting an embedded-sensor network for early warning of natural risks due to fast failures along railways;selective use of optimal image resolution for depth from multiple motions based on gradient scheme;and blood flow prediction and visualization within the aneurysm of the middle cerebral artery after surgical treatment.
the proceedings contain 81 papers. the topics discussed include: are buildings only instances? exploration in architectural style categories;geometry directed browser for personal photographs;heritage app: annotating ...
ISBN:
(纸本)9781450316606
the proceedings contain 81 papers. the topics discussed include: are buildings only instances? exploration in architectural style categories;geometry directed browser for personal photographs;heritage app: annotating images on mobile phones;content level access to digital library of India pages;large-scale statistical modeling of motion patterns: a Bayesian non-parametric approach;salient object detection using a fuzzy theoretic approach;a finite mixture model based on pair-copula construction of multivariate distributions and its application to color image segmentation;local appearance based robust tracking via sparse representation;semi-supervised multiple instance learning based domain adaptation for object detection;a grammar-based GUI for single view reconstruction;accelerating non-local denoising with a patch based dictionary;and viewpoint based mobile robotic exploration aiding object search in indoor environment.
Multispectral multifocus image fusion remains a challenging problem for the researchers in the computervision community. In this paper, we propose a novel solution to the above problem using guided filtering, steerab...
详细信息
ISBN:
(纸本)1595930361
Multispectral multifocus image fusion remains a challenging problem for the researchers in the computervision community. In this paper, we propose a novel solution to the above problem using guided filtering, steerable local frequency and an improved model of saliency. In the first place, promising fusion results are obtained through guided steerable local frequency maps. An accurate saliency map is developed next to further enhance these fusion results. Extensive experimentation on the visual, near infrared and thermal spectra clearly demonstrate the superiority of the proposed approach over some of the recently published works. Copyright 2014 ACM.
Duplication of images is a common occurrence in community based data sharing systems. An image of the same scene, residing as multiple copies in the system, introduces redundancy. this paper describes a novel techniqu...
详细信息
ISBN:
(纸本)1595930361
Duplication of images is a common occurrence in community based data sharing systems. An image of the same scene, residing as multiple copies in the system, introduces redundancy. this paper describes a novel technique to detect such submissions by matching the Speeded Up Robust Features (SURF) of a query image to the feature set of images in the database, which are pre-computed, dimensionality reduced, and indexed. First, a set of similar images is obtained withtheir feature key-point correspondences by computing homography. An occurrence of duplication is verified by statistical hypothesis testing, which considers the distribution obtained by inter-key-point Euclidean distance ratios between the corresponding key-points among the query and candidate images. Copyright 2014 ACM.
this paper investigates compression of depthimages-particularly, noisy depthimages-captured by depth sensors like Kinect. Our scheme is based on incrementally detecting planes in depthimages and then storing the pl...
详细信息
ISBN:
(纸本)1595930361
this paper investigates compression of depthimages-particularly, noisy depthimages-captured by depth sensors like Kinect. Our scheme is based on incrementally detecting planes in depthimages and then storing the plane parameters instead of the depth information of the pixels lying on the detected planes. Residuals are then computed and compressed using standard image compression techniques. Our technique incorporates the input error model for comprehensive and accurate plane detection. thereby, this accounts for the reliability of the input data in the compression scheme. the plane detection also accounts for edges. Experiments exhibit better image quality than standard compression techniques with smaller error. We additionally propose a novel error metric to evaluate compression of noisy depthimages. Copyright is held by the authors.
Jump Flooding is a method for propagating labels across a given plane from different seeds. It has been used to compute the discrete Voronoi tessellation of a given plane efficiently. We introduce a version of JFA, wh...
详细信息
ISBN:
(纸本)1595930361
Jump Flooding is a method for propagating labels across a given plane from different seeds. It has been used to compute the discrete Voronoi tessellation of a given plane efficiently. We introduce a version of JFA, which optimizes the number of pixels processed by computing only the faces of the Voronoi tessellation. the pixels in the interior of the Voronoi regions are not processed resulting in a 1-skeleton representation of the Voronoi tessellation in 2D and a 2-skeleton representation in 3D. We describe an implementation of this algorithm on a GPU using CUDA and demonstrate its performance benefits on multiple data sets. As an application of the proposed algorithm, we present a GPU based method for extraction of channel centerlines in biomolecules. the fast computation of the discrete Voronoi diagram is exploited to extract channels in molecular dynamics simulation trajectories on-the-fly, thereby supporting the interactive visual analysis of static and dynamic channel structures. Copyright is held by the authors.
暂无评论