Drucker, Steven M. and Zeltzer, David, CamDroid: a System for Implementing Intelligent Camera Control, Proceedings of 52 the 1995.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....potential for occlusion is not addressed and when the constraints are not satisfiable the approximate solution can always be found, but may well be useless. 2.2.6. 2 CINEMA A generalised version of the approach of Blinn was developed by Drucker [Drucker et al. 1992] Drucker and Zeltzer, 1994] [Drucker and Zeltzer, 1995] [Drucker, 1994] In Drucker s system the desired image is defined by two sets of functions the first are constraints that must be met and the second are objectives which have to be minimised. The resulting constrained optimisation problem is passed to an optimisation package called C Code for ....

Drucker, S. M. and Zeltzer, D. (1995). Camdroid: a system for implementing intelligent camera control. In SIGGRAPH Symposium on Interactive 3D Graphics, volume 28, pages 196--199. Computer Graphics.

....to rapidly cover distance[21] These methods do not assist the user in the actual navigation to a particular goal. Another collection of papers studies the computation of e#ective fixed camera positions to assist the user in performing certain (manipulation) tasks, the so called shot systems [4, 8, 11]. To achieve this a collection of constraints is often derived which is then solved. These systems do not plan obstacle avoiding motions. Most previous work on planning camera motions is directed toward systems in which the camera must follow an object (like in third person games) A distinction ....

S.M. Drucker, D. Zeltzer, CamDroid: A system for implementing intelligent camera control, In: P. Hanrahan and J.Winget (Eds), SIGGRAPH Symposium on Interactive 3D Graphics, 1995, pp. 139--144.

....to lists, requires prior specific permission and or a fee. IUI 01, January 14 17, 2001, Sante Fe, New Mexico. Copyright 2001 ACM 1 58113 325 1 01 0001. 5.00. Intelligent visualization researchers have built systems for automatic explanation generation, intelligent tutoring, and other tasks [10, 20, 8, 9, 14, 3, 4], relying on many of the same sources we use. The requirements for our work differ in some ways from these efforts, however: the assistant and the user observe an external process, each able to guide it but without complete control; camera manipulation is viewed as a means of explicit ....

Drucker, S., and Zelter, D. 1995. Camdroid: A system for implementing intelligent camera control. Proceedings 1995 Symposium on Interactive 3D Graphics, 139-144.

....The CUE Video effort at IBM Research [22] is notable as it uses a combination of vision and signal processing techniques to address all of these questions. Another related set of research uses cinematography knowledge to create animated presentations. For example, Drucker and Zeltzer [5] cast the camera motion planning problem as a constrained optimization, He et al. 9] choose camera motions for virtual dialog scenes, and Bares and Lester [1] automatically control cameras in virtual worlds. 3 Available Building Blocks In order to produce a video, a videographer uses their ....

S. Drucker and D. Zeltzer. Camdroid: A system for implementing intelligent camera control. 1995.

....pictures look, as well as the temporal aspects, to create motions that do not confuse or sicken the viewer. Ultimately, a virtual videography system may encode heuristics that define the art of photography and cinematography. This has been explored in a constraint based framework by Drucker [7] in the context of 3D virtual environments. Initially, our virtual videography experiments use simple algorithms to frame important elements, and we use filtering to avoid the generation of jittery motions. Implementing the filtering by fitting known good movement patterns, such as ease in out, ....

S. Drucker and D. Zeltzer. Camdroid: A system for implementing intelligent camera control. 1995 Symposium on Interactive 3D Graphics, pages 139--144, April 1995.

....positions. In this paper, by contrast, we are concerned with real time camera placement as the interactively controlled action proceeds. A number of other systems concentrate on finding the best camera placement when interactive tasks are performed [8, 14, 16] In particular, Drucker et al. [4, 5, 6] show how to set up the optimal camera positions for individual shots by solving small constrained optimization problems. For efficiency reasons, in the real time setting we select shots from a small set of possible camera specifications so that camera positions can be computed using closed form ....

....due to unexpected occlusions, or they may miss a critical action due to minimum length shot constraints. Some of these issues can be resolved by redesigning the idioms in the current structure. We are also looking into incorporating simple constraint solvers, such as the ones proposed by Drucker [4, 5, 6] and Gleicher and Witkin [8] In addition, we would like to expand the input to the VC to include such ancillary information as the emotional state of the scene or of individual actors. For example, if the situation is tense, faster cuts might be made, or if one actor is scared, the camera might ....

Steven M. Drucker and David Zelter. CamDroid: A system for implementing intelligent camera control. In Michael Zyda, editor, Computer Graphics (1995 Symposium on Interactive 3D Graphics), volume 28, pages 139--144, April 1995.

.... interactivity [11, 6] 2) requires users to directly control low level camera positioning and orientation parameters [10, 13, 18, 24] which is problematic when users must perform complex tasks while simultaneously issuing camera control commands, or (3) provides automated realtime camera control [4, 9, 5, 21] but does not take into account the inherent intentionality of users activities as they perform complex navigational and manipulative tasks. To address this problem, we have developed the cinematic taskmodeling framework for automated realtime task sensitive camera control in 3D environments. ....

....the direct control approach [10, 13, 18, 24] camera systems require users to directly control low level camera positioning and orientation parameters. This is problematic when users must perform complex tasks while simultaneously issuing camera control commands. In the realtime planning approach [4, 5, 9, 21], camera systems dynamically plan camera position and orientation. While this approach comes closest to delivering the necessary functionalities required of dynamic 3D learning environments, previous work on automated 3D camera planning has not taken into account the inherent intentionality of ....

[Article contains additional citation context not shown here]

S. Drucker and D. Zeltzer. CamDroid: A system for implementing intelligent camera control. In Proceedings of the 1995 Symposium on Interactive 3D Graphics, pages 139--144, 1995.

....respond by selecting one of the pre speci#ed relative camera positions, e.g. placing the camera at a point displaced by a given vector from the subject#s#. In contrast, the constraint satisfaction approachto automated cinematography casts camera planning as a constraint satisfaction problem #Drucker Zeltzer 1995#. Given a request to view particular subjects of interest and a speci#cation of how each should be viewed, the CamDroid constraint solver attempts to #nd a solution. In general, the constraint based approach offers signi#cantly greater #exibility than alternate approaches. However, this initial ....

Drucker, S., and Zeltzer, D. 1995. CamDroid: A system for implementing intelligent camera control. In Proceedings of the 1995 Symposium on Interactive 3D Graphics, 139#144.

....is used to compute the time derivates of the actual viewing parameters which allows for controls to be defined independently of the underlying view parametrization. Many other researchers have addressed research issues associated with view definitions and in particular viewpoint movements [7, 14, 8, 2, 3, 6, 11]. However, most of the developed techniques are heavily application dependent. They do not allow a user to interactively define multiple views on a 3D scene in an intuitive, graphical manner such that each view can be configured and parametrized according to the user s needs. 2.2 The CSE and PGO ....

S.M. Drucker and D. Zeltzer. CamDroid: A system for implementing intelligent camera control. In P. Han- REFERENCES 8

.... has taken advantage of cinematography heuristics to produce systems for automatic explantation generation, intelligent tutoring, and other tasks (Feiner McKeown 1991; Smith Bates 1989; Karp Feiner 1990; Seligmann Feiner 1991; Gleicher Witkin 1992; Phillips, Badler, Granieri 1992; Drucker Zelter 1994; 1995; Christianson et al. 1996; He, Cohen, Salesin 1996; Bares Lester 1997; 1999) Although some of these systems present a direct manipulation interface to the user, the camera is not considered a method for communication; instead, camera planning is simply used to orient the user s perspective ....

Drucker, S., and Zelter, D. 1995. Camdroid: A system for implementing intelligent camera control. In Proceedings of the 1995 Symposium on Interactive 3D Graphics, 139--144.

....visualize particular subjects of interest and specifies how each should be viewed, a constraint solver attempts to find a solution camera placement. CAMDROID supports a broad and powerful set of camera constraints, but employs a numerical constraint solver that is subject to local minima failures [6, 7]. In our prior work, a task model was used to direct a real time constraint based camera planner to provide views of a learner s avatar [3] Neither of these two constraint based systems provided a systematic solution for handling constraint failures. Automatic camera control assistants vary ....

....Interface Constraint Solver Formulating Camera Constraints A user s visualization goal is expressed as a constraint problem consisting of a number of constraints on the subject(s) to be viewed. The Constraint Solver supports four types of constraints (a subset of Drucker s constraints [6, 7]) Each constraint can be applied to any subject to be viewed by the camera and includes a relative priority and a marker indicating whether that constraint can be relaxed. Vantage Angle: Indicates the permissible range of relative orientations between the camera and the subject, e.g. which ....

[Article contains additional citation context not shown here]

Steven Drucker and David Zeltzer. CamDroid: A system for implementing intelligent camera control. In Proceedings of the 1995 Symposium on Interactive 3D Graphics, pages 139-144, 1995.

....CAVE human play along 3D immersive Table 1. Overview of different visualization modes and their characteristics mated camera system that tracks the ball and switches between camera positions according to a user definable algorithm. This mode thus is an approximation of the standard TV coverage [5, 9]. The second mode is implemented on a workbench 2 and provides a 3D overview of the soccer match. Its main feature is that visualization is miniaturized to allow a human to overview the state (without a predefined viewing angle) The goal here is to allow a coach to steer the game and give ....

S. M. Drucker and D. Zeltzer. CamDroid: A system for implementing intelligent camera control. In 1995 Symposium on Interactive 3D Graphics, pages 139--144, Apr. 1995.

....true in desktop 3D environments where the user realizes that a slightly different viewpoint would useful while performing another task. Rather than interrupt the task, change the viewpoint, and resume the task the user can rely on the system determining the most useful viewpoint. Phillips][Drucker][He] have all experimented with different techniques for automatically changing the user s viewpoint to provide interesting or more useful views, and some video games control the user s viewpoint to free the user for other tasks. For example, when the user is moving an object around the system ....

Steven M. Drucker and David Zeltzer. CamDroid: A System for Implementing Intelligent Camera Control. Proceedings of the 1995 Symposium on Interactive 3D Graphics, pages 139-144.

....Workstations Cave (Stockholm) Cave (Amsterdam) Workbench (Delft) Soccer game Visualization Figure 2. Interactive and collaborative visualization of a soccer match between camera positions according to a user definable algorithm. This mode thus is an approximation of the standard TV coverage [3, 7]. The second mode is implemented on a workbench 2 and provides a 3D overview of the soccer match. Its main feature is that visualization is miniaturized to allow a human to overview the state (without a predefined viewing angle) The goal here is to allow a coach to steer the game and give ....

S. M. Drucker and D. Zeltzer. CamDroid: A system for implementing intelligent camera control. In 1995 Symposium on Interactive 3D Graphics, pages 139--144, Apr. 1995.

No context found.

Drucker, Steven M. and Zeltzer, David, CamDroid: a System for Implementing Intelligent Camera Control, Proceedings of 52 the 1995.

No context found.

S. M. Drucker and D. Zeltzer. CamDroid: A system for implementing intelligent camera control. In Computer Graphics (SIGGRAPH '95 Proceedings), 1996.

No context found.

Steven Drucker and David Zeltzer. Camdroid: A System for Implementing Intelligent Camera Control. 1995 Symposium on Interactive 3D Graphics, Monterey, CA, USA, 1995.

No context found.

S. M. Drucker and D. Zeltzer. CamDroid: A system for implementing intelligent camera control. In P. Hanrahan and J. Winget, editors, 1995 Symposium on Interactive 3D Graphics, pages 139--144. ACM SIGGRAPH, Apr. 1995. ISBN 0-89791-736-7. 2

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