G. D. Hager and K. Toyama, "The "X-vision" system: A general purpose substrate for real-time vision applications," Comp. Vision, Image Understanding., vol. 69, pp. 23--27, January 1998.  Home/Search   Document Not in Database   Summary   Related Articles   Check

No context found.

G. D. Hager and K. Toyama, "The "X-vision" system: A general purpose substrate for real-time vision applications," Comp. Vision, Image Understanding., vol. 69, pp. 23--27, January 1998.

....At the same time, systems defined by Frob are amenable to formal reasoning and program transformation. We are currently working to broaden our experience with FRP in the robotics domain in a number of ways. First, as noted above we are developing FVision, based on the existing XVision [14] system, to support vision research and integrated vision and robotics. We also plan to develop an environment for robotic hand eye coordination to test our ideas in a more challenging run time environment. Finally, we are using Frob to control a team of robots in the Robocup soccer competition. ....

G. D. Hager and K. Toyama, "The "XVision" system: A general purpose substrate for real-time vision applications, " Comp. Vision, Image Understanding., vol. 69, pp. 23--27, Jan. 1998.

....into this framework. ffl Systems defined by Frob are amenable to formal reasoning and program transformation. We are currently working to broaden our experience with FRP in the robotics domain in a number of ways. First, as noted above we are developing FVision, based on the existing XVision [8] system, to support vision research and integrated vision and robotics. We also plan to develop an environment for robotic hand eye coordination to test our ideas in a more challenging run time environment. Due to space limitations, we have been forced to omit some details of the coded examples. ....

G. D. Hager and K. Toyama. The "XVision" system: A general purpose substrate for real-time vision applications. Comp. Vision, Image Understanding., 69(1):23--27, January 1998.

....that is at a critical juncture. Inexpensive cameras, digitizers, and highperformance video devices are now plentiful, and the processing power of most PC s and workstations has reached the point where they can perform many image processing functions which historically required specialized hardware [7]. Software exploiting vision has not, however, advanced at a comparable rate. We assert that this is not due to a lack of algorithms or computing power, but rather that little is yet known about the effective software abstractions and tools in this domain. There have been several attempts to build ....

G. D. Hager and K. Toyama. The "XVision" system: A general purpose substrate for real-time vision applications. Comp. Vision, Image Understanding. , 69(1):23--27, January 1998.

....vision based motion control modules which can be used by non experts. For a review of visualservoing techniques, see [5] We have already developed a real time vision system, which we call XVision, that allows users to perform fast feature tracking based on gradient edges and texture patches [13]. The XVision system comprises several modular software components which can be combined to track objects of varying complexity, ranging from a simple line segment to human faces. This article describes a similar modular software system we are developing for robot control tasks: Servomatic allows ....

....it can be shown that p i Delta l i is the distance between the point and the projection of the line in the image plane. Since L v is not defined when l 1 and l 2 are parallel, any epipolar plane is a singular region for stereo line projection. Estimates of a line L from l are written L(l) See [13] for a more detailed exposition on point and line representation. 3 Positioning Skills Next, we define two primitive positioning operations utilizing point and line features. We refer to [12] for an in depth description of these skills and additional regulators for line to point positioning and ....

[Article contains additional citation context not shown here]

G. D. Hager. The "X-vision" system: A general purpose substrate for real-time vision-based robotics. In Proceedings of the Workshop on Vision for Robots, pages 56--63, 1995. Also available as Yale CS-RR1078.

....search a rectangular region of an image about a nominally predicted location. The search space for such a tracker may be restricted to a predefined size in order to assure reasonable real time performance. On the other hand, some window based line segment trackers (such as those described in [10]) are designed to find lines within a band of orientations, lengths, and image locations. In order to connect a point feature tracker to a line tracker, some translation and partitioning of output sets is required. Selectors also focus the attention of the system by favoring certain output sets ....

....other lower level algorithms can take over. The following are a few examples of trackers and selectors: Polygon Tracker: A feature based polygon tracker tracks polygonal shapes as multiple corners, which are in turn tracked as two intersecting line segments (a rectangle tracker is described in [10]) The polygonal tracker will narrow a compact set of input configurations, into an smaller set of configurations which corresponds to the image projection of the polygon together with some allowance for noise. Mistracking is detected when the line segments of adjacent corners are out of ....

[Article contains additional citation context not shown here]

G. D. Hager. The "X-vision" system: A general purpose substrate for real-time visionbased robotics. In Proceedings of the Workshop on Vision for Robots, pages 56--63, 1995. Also available as Yale CS-RR-1078.

....of images at camera rate. However not all pixels in the image are of interest, and computation time can be greatly reduced if only a small region around each image feature is processed. Thus, a promising technique for making vision cheap and tractable is to use window based tracking techniques [54,70,71]. Windowbased methods have several advantages, among them; computational simplicity, little requirement for special hardware, and easy reconfiguration for different applications. We note, however, that initialization of window or region based systems typically presupposes an automated or ....

....over time, but the brightness of the opening and the brightness of the surrounding region are roughly constant, a circular template could be localized using SSD methods augmented with brightness and contrast parameters. In this case, 59) must also include parameters for scaling and aspect ratio [70]. 3. The opening could be selected in an initial image, and subsequently located using SSD methods. This differs from the previous method in that this calculation does not compute the center of the opening, only its correlation with the starting image. Although useful for servoing a camera to ....

[Article contains additional citation context not shown here]

G. D. Hager, "The "X-Vision" system: A general purpose substrate for real-time vision-based robotics." Submitted to the 1995 Workshop on Vision for Robotics, Feb. 1995.

.... has also proliferated [2, 4, 10, 12, 27, 28, 32, 34, 39, 43] Some of this work considers task composition and discrete event systems to link active vision modules, e.g. 29, 30, 14] In our own work, we have begun the development of modular systems for hand eye coordination and visual tracking [19, 21, 25, 26]. Our work differs from other work in several ways. First, we emphasize a modular approach where task primitives are defined for simple subtasks that occur repeatedly in the domain. These primitives are composed to perform complex task via consistent design principles, where composition can occur ....

....control architecture described above, we have developed a corresponding set of visual tracking primitives. This system, called XVision, also incorporates a notion of primitive tracking composition similar in spirit to parallel composition. Here, we highlight XVision s fundamental features; [17, 19, 36, 35] contain more detailed descriptions. Every tracking primitive returns a parameter vector the vector of parameters it observes. Tracking algorithms follow a cyclic process consisting of three phases: In the predict phase, an initial guess is made for the parameter vector of the primitive. The ....

G. D. Hager and K. Toyama. The "XVision" system: A general purpose substrate for real-time vision applications. Comp. Vision, Image Understanding., 69(1):23-- 27, January 1998.

....simple sensor based control loops visual, force, or position based servoing suffice for most of the simple manipulations to be performed by the system. In particular, we plan to rely strongly on the visual tracking and control concepts developed by Hager (Hager 1997; Hager and Toyama 1996; Hager and Toyama 1995). Briefly, the behavior of the system is driven by the system state model and the target state model. Intuitively, the target state model specifies values for specific sensor features which correspond to a correctly performed task or subtask. These values may be physical values or, in some cases, ....

....4.1 An Overview The vision processing to be performed by the system can be divided into two different modes of operation: visual tracking and visual search. Visual tracking, used primarily for feature based visual servoing, is performed by the XVision package (Hager and Toyama 1996; Hager and Toyama 1995). XVision is a portable, software based processing system optimized for local feature tracking on commodity computing hardware. It is organized around the idea of decomposing tracking problems into a small set of highly optimized image level features. These features include unstructured blobs, ....

Hager, G. D. and K. Toyama (1995). The "Xvision " system: A general purpose substrate for real-time vision applications. DCS RR-1078, Yale University. To Appear in Comp. Vis. Image Understanding.

....Although vision provides a huge amount of data, we quickly focus attention on small portions of the image which are easily distinguished from their local (in the image) surroundings, and track these patches through image sequences. Visual tracking of this type has proven to be simple to perform [2, 10], yet it is robust and it reduces image information to a time history of a small set of feature locations. Consequently, the set of nominal robot paths in our system is represented in terms of the image trajectories of tracked features. During subsequent navigation, the image motion of observed ....

....the system with an independent pan axis for the camera would achieve nearly the same results. We assume that the robot has already acquired data about a set of nominal paths through the environment. Here we briefly summarize the important aspects of that representation and refer the reader to [26, 8, 10] for more details on the techniques used to acquire it. The set of nominal paths is represented as a directed graph (the map) based on the recorded visual trajectories of tracked features which we call markers. We represent the trajectory of marker i as a function m i (t) b i t e i ; where t ....

G. D. Hager and K. Toyama. The "XVision" system: A general purpose substrate for real-time vision applications. Comp. Vision, Image Understanding., 69(1):23--27, January 1998.

....of the polygon which would be consistent with the object projecting the tracked edge onto the camera image. ffl A feature based polygon tracker tracks polygonal shapes as multiple corners, which are in turn tracked as two intersecting line segments (a rectangle tracker is described in [19]) The polygonal tracker will narrow a compact set of input configurations, into an smaller set of configurations which corresponds to the image projection of the polygon together with some allowance for noise. Mistracking is detected when the line segments of adjacent corners are out of alignment ....

G. D. Hager. The "X-Vision" system: A general purpose substrate for real-time vision-based robotics. In Proceedings of the Workshop on Vision for Robots, pages 56--63, 1995. Also available as Yale CS-RR-1078.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC