Results 1 - 10
of
52
Optimal placement of cameras in floorplans to satisfy task requirements and cost constraints
- In Proc. of OMNIVIS Workshop
, 2004
"... Abstract. In many multi-camera vision systems the effect of camera locations on the task-specific quality of service is ignored. Researchers in Computational Geometry have proposed elegant solutions for some sensor location problem classes. Unfortunately, these solutions utilize unrealistic assumpti ..."
Abstract
-
Cited by 27 (0 self)
- Add to MetaCart
(Show Context)
Abstract. In many multi-camera vision systems the effect of camera locations on the task-specific quality of service is ignored. Researchers in Computational Geometry have proposed elegant solutions for some sensor location problem classes. Unfortunately, these solutions utilize unrealistic assumptions about the cameras’ capabilities that make these algorithms unsuitable for many real-world computer vision applications. In this paper, the general camera placement problem is first defined with assumptions that are more consistent with the capabilities of realworld cameras. Given a floorplan to be observed, the problem is to efficiently compute a camera layout such that certain task-specific constraints are met and with minimal camera setup cost. A solution to this problem is obtained via binary optimization over a discrete problem space. In preliminary experiments the performance of the system is demonstrated with two different practical experiments on a real floorplan. 1
Robust tracking in a camera network: A multiobjective optimization framework
- Selected Topics in Signal Processing: Special Issue on Distributed Processing in Vision Networks, 2(4):582 – 596
, 2008
"... We address the problem of tracking multiple people in a network of non-overlapping cameras. This introduces certain challenges that are unique to this particular application scenario, in addition to existing challenges in tracking like pose and illumination variations, occlusion, clutter and sensor ..."
Abstract
-
Cited by 16 (3 self)
- Add to MetaCart
(Show Context)
We address the problem of tracking multiple people in a network of non-overlapping cameras. This introduces certain challenges that are unique to this particular application scenario, in addition to existing challenges in tracking like pose and illumination variations, occlusion, clutter and sensor noise. For this purpose, we propose a novel multi-objective optimization framework by combining short term feature correspondences across the cameras with long-term feature dependency models. The overall solution strategy involves adapting the similarities between features observed at different cameras based on the long-term models and finding the stochastically optimal path for each person. For modeling the long-term interdependence of the features over space and time, we propose a novel method based on discriminant analysis models. The entire process allows us to adaptively evolve the feature correspondences by observing the system performance over a time window, and correct for errors in the similarity estimations. We show results on data collected by two large camera networks. These experiments prove that incorporation of the long-term models enable us to hold tracks of objects over extended periods of time, including situations where there are large “blind ” areas. The proposed approach is implemented by distributing the processing over the entire network.
Collaborative sensing in a distributed ptz camera network,” Image Processing
- IEEE Transactions on
, 2012
"... Abstract—The performance of dynamic scene algorithms often suffers because of the inability to effectively acquire features on the targets, particularly when they are distributed over a wide field of view. In this paper, we propose an integrated analysis and control framework for a pan, tilt, zoom ( ..."
Abstract
-
Cited by 14 (1 self)
- Add to MetaCart
(Show Context)
Abstract—The performance of dynamic scene algorithms often suffers because of the inability to effectively acquire features on the targets, particularly when they are distributed over a wide field of view. In this paper, we propose an integrated analysis and control framework for a pan, tilt, zoom (PTZ) camera network in order to maximize various scene understanding performance criteria (e.g., tracking accuracy, best shot, and image resolution) through dynamic camera-to-target assignment and efficient feature acquisition. Moreover, we consider the situation where processing is distributed across the network since it is often unrealistic to have all the image data at a central location. In such situations, the cameras, although autonomous, must collaborate among themselves because each camera’s PTZ parameter entails constraints on the others. Motivated by recent work in cooperative control of sensor networks, we propose a distributed optimization strategy, which can be modeled as a game involving the cameras and targets. The cameras gain by reducing the error covariance of the tracked targets or through higher resolution feature acquisition, which, however, comes at the risk of losing the dynamic target. Through the optimization of this reward-versus-risk tradeoff, we are able to control the PTZ parameters of the cameras and assign them to targets dynamically. The tracks, upon which the control algorithm is dependent, are obtained through a consensus estimation algorithm whereby cameras can arrive at a consensus on the state of each target through a negotiation strategy. We analyze the performance of this collaborative sensing strategy in active camera networks in a simulation environment, as well as a real-life camera network. Index Terms—Camera networks, cooperative camera control, distributed estimation, game theory, video analysis. I.
Distributed perimeter patrolling and tracking for camera networks
- In Decision and Control (CDC), 2010 49th IEEE Conference on
"... Abstract — In this work, we propose a distributed control strategy for perimeter patrolling and target tracking in a multi-camera videosurveillance system with communication, resources and speed constraints. These cameras are required to monitor a perimeter and share common portions of this perimete ..."
Abstract
-
Cited by 13 (2 self)
- Add to MetaCart
(Show Context)
Abstract — In this work, we propose a distributed control strategy for perimeter patrolling and target tracking in a multi-camera videosurveillance system with communication, resources and speed constraints. These cameras are required to monitor a perimeter and share common portions of this perimeter to allow redundant coverage. We propose an algo-rithm that is able to find the global patrolling strategy only through local asynchronous communication and coordination of neighboring cameras even in presence of physical limits of each camera visibility area. The algorithm converges to an optimal solution, and its distribuited implementation is obtainet through an electric circuit analogy. The proposed system also includes a Kalman-based filter for each camera to track moving targets within its areas of competence, and a distributed coordination scheme for target hand-off between different cameras that guarantees target locking at all times. Numerical simulations are provided to test the proposed algorithms.
Sensor planning for automated and persistent object tracking with multiple cameras
- IEEE Conf. on Computer Vision and Pattern Recognition
, 2008
"... Most existing camera placement algorithms focus on coverage and/or visibility analysis, which ensures that the object of interest is visible in the camera’s field of view (FOV). However, visibility, a fundamental requirement of object tracking, is insufficient for persistent and automated tracking. ..."
Abstract
-
Cited by 11 (3 self)
- Add to MetaCart
(Show Context)
Most existing camera placement algorithms focus on coverage and/or visibility analysis, which ensures that the object of interest is visible in the camera’s field of view (FOV). However, visibility, a fundamental requirement of object tracking, is insufficient for persistent and automated tracking. In such applications, a continuous and consistently labeled trajectory of the same object should be maintained across different cameras ’ views. Therefore, a sufficient overlap between the cameras ’ FOVs should be secured so that camera handoff can be executed successfully and automatically before the object of interest becomes untraceable or unidentifiable. The proposed sensor planning method improves existing algorithms by adding handoff rate analysis, which preserves necessary overlapped FOVs for an optimal handoff success rate. In addition, special considerations such as resolution and frontal view requirements are addressed using two approaches: direct constraint and adaptive weight. The resulting camera placement is compared with a reference algorithm by Erdem and Sclaroff. Significantly improved handoff success rate and frontal view percentage are illustrated via experiments using typical office floor plans. 1.
Distributed Camera Networks -- Integrated sensing and analysis for wide-area scene understanding
, 2011
"... ..."
presented at
- Budapest 2001 High Energy Physics Conference”, hep-ph/0112026
"... I am submitting herewith a thesis written by Marcus James Jackson entitled “Wide-Area ..."
Abstract
-
Cited by 9 (1 self)
- Add to MetaCart
I am submitting herewith a thesis written by Marcus James Jackson entitled “Wide-Area
A COMPLETE ALGORITHM FOR SEARCHLIGHT SCHEDULING
, 2011
"... This article develops an algorithm for a group of guards statically positioned in a nonconvex polygonal environment with holes. Each guard possesses a single searchlight, a ray sensor which can rotate about the guard’s position but cannot penetrate the boundary of the environment. A point is detecte ..."
Abstract
-
Cited by 7 (1 self)
- Add to MetaCart
This article develops an algorithm for a group of guards statically positioned in a nonconvex polygonal environment with holes. Each guard possesses a single searchlight, a ray sensor which can rotate about the guard’s position but cannot penetrate the boundary of the environment. A point is detected by a searchlight if and only if the point is on the ray at some instant. Targets are points which move arbitrarily fast. The objective of the proposed algorithm is to compute a schedule to rotate a set of searchlights in such a way that any target in an environment will necessarily be detected in finite time. This is known as the Searchlight Scheduling Problem and was described originally in 1990 by Sugihara et al. We take an approach known as exact cell decomposition in the motion planning literature. The algorithm operates by decomposing the searchlights ’ joint configuration space and the environment, and then by constructing a so-called information graph. Searching the information graph for a path between desired states yields a search schedule. We also introduce a new problem called the φ-Searchlight Scheduling Problem in which φ-searchlights sense not just along a ray, but over a finite field of view. We show that our results for searchlight scheduling can be directly extended for φ-searchlight scheduling. Proofs of completeness, complexity bounds, and computed examples are presented.
An Exact and Efficient Algorithm for the Orthogonal Art Gallery Problem
"... In this paper, we propose an exact algorithm to solve the Orthogonal Art Gallery problem in which guards can only be placed on the vertices of the polygon P representing the gallery. Our approach is based on a discretization of P into a finite set of points in its interior. The algorithm repeat-edly ..."
Abstract
-
Cited by 7 (2 self)
- Add to MetaCart
(Show Context)
In this paper, we propose an exact algorithm to solve the Orthogonal Art Gallery problem in which guards can only be placed on the vertices of the polygon P representing the gallery. Our approach is based on a discretization of P into a finite set of points in its interior. The algorithm repeat-edly solves an instance of the Set Cover problem obtaining a minimum set Z of vertices of P that can view all points in the current discretization. Whenever P is completely vis-ible from Z, the algorithm halts; otherwise, the discretiza-tion is refined and another iteration takes place. We estab-lish that the algorithm always converges to an optimal solu-tion by presenting a worst case analysis of the number of it-erations that could be effected. Even though these could the-oretically reach O(n4), our computational experiments re-veal that, in practice, they are linear in n and, for n ≤ 200, they actually remain less than three in almost all instances. Furthermore, the low number of points in the initial dis-cretization, O(n2), compared to the possible O(n4) atomic visibility polygons, renders much shorter total execution times. Optimal solutions found for different classes of in-stances of polygons with up to 200 vertices are also de-scribed. 1. Introduction and Related
