Evaluating motion graphs for character navigation

Reitsma, Paul S. A.; Pollard, Nancy S.

doi:10.1145/1028523.1028536

Cited by 41 publications

(35 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This result is anticipated as we know that the character's minimum turning radius is quite large. Inspired by the work of Reitsma and Pollard [2004], we used a discrete, brute force method to embed our parametric motion graph in the environment in hopes of better understanding this problem. This embedding made it clear that our character could easily meet location constraints within a reasonable radius but that for most locations, there were only a few orientations that the character could be in when they arrived.…”

Section: Target Directed Controlmentioning

confidence: 99%

Parametric motion graphs

Heck

Gleicher

2007

Proceedings of the 2007 Symposium on Interactive 3D Graphics and Games

175

120

View full text Add to dashboard Cite

Figure 1: An interactively controllable walking character using parametric motion graphs to smoothly move through an environment. The character is turning around to walk in the user-requested travel direction, depicted by the red arrow on the ground. AbstractIn this paper, we present an example-based motion synthesis technique that generates continuous streams of high-fidelity, controllable motion for interactive applications, such as video games. Our method uses a new data structure called a parametric motion graph to describe valid ways of generating linear blend transitions between motion clips dynamically generated through parametric synthesis in realtime. Our system specifically uses blending-based parametric synthesis to accurately generate any motion clip from an entire space of motions by blending together examples from that space. The key to our technique is using sampling methods to identify and represent good transitions between these spaces of motion parameterized by a continuously valued parameter. This approach allows parametric motion graphs to be constructed with little user effort. Because parametric motion graphs organize all motions of a particular type, such as reaching to different locations on a shelf, using a single, parameterized graph node, they are highly structured, facilitating fast decision-making for interactive character control. We have successfully created interactive characters that perform sequences of requested actions, such as cartwheeling or punching.

show abstract

Section: Target Directed Controlmentioning

confidence: 99%

Parametric motion graphs

Heck

Gleicher

2007

Proceedings of the 2007 Symposium on Interactive 3D Graphics and Games

175

120

View full text Add to dashboard Cite

show abstract

“…However, the quality of the results depends largely on the size (Treuille et al, 2007) (only a few minutes (Safonova and Hodgins, 2007)) as well as the quality of the motion database used to construct the motion graph (the appropriateness of the database (Chai and Hodgins, 2007), whereby once again motion capture data is used as a motion prior while matching constraints are specified by the user), together with the "good", and/or automatic, connectivity of the motion graph (Ikemoto et al, 2007). Furthermore, the idea of evaluating motion graphs is outlined by Reitsma and Pollard (Reitsma and Pollard, 2004), in order to identify how much motion data should be included in the motion graph, in addition to assessing the character's capabilities while comparing motion datasets for the improvement of the motion graph.…”

Section: Discussionmentioning

confidence: 99%

Human Figure Animation: A Historical Perspective

Chrysanthou¹,

Loizides²

2008

ECMS 2008 Proceedings Edited By: L. S. Louca, Y. Chrysanthou, Z. Oplatkova, K. Al-Begain

View full text Add to dashboard Cite

Computer Animation, has advanced significantly as well as spectacularly over the last two-to-three decades. The interesting field has innovated tremendously moving from first-generation i.e. purely geometric models where motion was developed using kinematic techniques, to secondgeneration i.e. physical-based models (direct versus inverse dynamics and hybrid systems), towards the development of the third-generation i.e. autonomous behaviour through learning and perception. The simulation and animation of a variety of real-world objects appear with stunning realism. However, the main issue here is not on what we have achieved but rather, based on what we have achieved, what is to follow. The emerging results are expected immensely by researchers in the field, worldwide. The paper delivers a survey on human figure animation and in particular a classification framework of the work done on the control of locomotion.

show abstract

“…Directly capturing transitions between behaviors produces higher quality motion data than interpolating motion segments between behaviors. Assessing the final quality of a motion dataset is a difficult problem; Reitsma and Pollard [75] describe an algorithm for embedding motion capture data into a target environment and calculating path quality and environment coverage. In practice, it is possible for an experienced technician to evaluate the dataset by visually observing a synthetic character animated using the data and noting dis continuities in the character's motion.…”

Section: Data Collectionmentioning

confidence: 99%

Activity Recognition for Agent Teams

Sycara

Sukthankar

2007

View full text Add to dashboard Cite

Proficient teams can accomplish goals that would not otherwise be achievable by groups of uncoordinated individuals. This thesis addresses the problem of analyzing team activities from external observations and prior knowledge of the team's behavior patterns. There are three general classes of recognition cues that are potentially valuable for team activity/plan recognition: (1) spatial relationships between team members and/or physical landmarks that stay fixed over a period of time; (2) temporal dependencies between behaviors in a plan or between actions in a behavior; (3) coordination constraints between agents and the actions that they are performing. This thesis examines how to leverage available spatial, temporal, and coordination cues to perform offline multi-agent activity/plan recognition for teams with dynamic membership.In physical domains (military, athletic, or robotic), team behaviors often have an observable spatio-temporal structure, defined by the relative physical positions of team members and their relation to static landmarks. We suggest that this structure, along with temporal dependencies and coordination constraints defined by a team plan library, can be exploited to perform behavior recognition on traces of agent activity over time, even in the presence of uninvolved agents. Unlike prior work in team plan recognition where it is assumed that team membership stays constant over time, this thesis addresses the novel problem of recovering agent-to-team assignment for team tasks where team composition, the mapping of agents into teams, changes over time; this allows the analysis of more complicated tasks in which agents must periodically divide into subteams.This thesis makes four main contributions: (1) an efficient and robust technique for formation identification based on spatial relationships; (2) a new algorithm for simultaneously determining team membership and performing behavior recognition on spatio-temporal traces with dynamic team membership; (3) a general pruning technique based on coordination cues that improves the efficiency of plan recognition for dynamic teams; (4) methods for identifying player policies in team games that lack strong spatial, temporal, and coordination dependencies.ii

show abstract

Evaluating motion graphs for character navigation

Abstract: Abstract

Cited by 41 publications

References 49 publications

Parametric motion graphs

Parametric motion graphs

Human Figure Animation: A Historical Perspective

Activity Recognition for Agent Teams

Contact Info

Product

Resources

About