Presentation is loading. Please wait.

Presentation is loading. Please wait.

An Evaluation of a Cost Metric for Selecting Transitions between Motion Segments EG SCA ’03 Speaker: Alvin Date: 4 October 2004.

Similar presentations


Presentation on theme: "An Evaluation of a Cost Metric for Selecting Transitions between Motion Segments EG SCA ’03 Speaker: Alvin Date: 4 October 2004."— Presentation transcript:

1 An Evaluation of a Cost Metric for Selecting Transitions between Motion Segments EG SCA ’03 Speaker: Alvin Date: 4 October 2004

2 Alivn/GAME Lab./CSIE/NDHUAn Evaluation of a Cost Metric for Selecting Transitions between Motion Segments 2 Outline Introduction Cost Function Optimization Evaluation  Cross-Validation  User Study Evaluation Form Conclusion & Future Works

3 Alivn/GAME Lab./CSIE/NDHUAn Evaluation of a Cost Metric for Selecting Transitions between Motion Segments 3 Problem Define Goal  Design rich behaviors.  Minimize the amount of manual intervention. Input  MoCap data Methods  Use an underlying probabilistic model.  Re-order the original motion. Key  Proper selection of transition points.

4 Alivn/GAME Lab./CSIE/NDHUAn Evaluation of a Cost Metric for Selecting Transitions between Motion Segments 4 Introduction Transition Points  The discontinuities in the motion stream. Distance Function  The cost of transitioning from one frame to another.  Parameterized though user-selected weights. Contribution  Produce a set of optimized weights that select good transitions.

5 Alivn/GAME Lab./CSIE/NDHUAn Evaluation of a Cost Metric for Selecting Transitions between Motion Segments 5 Cost Function Lee  d(v i,v j ) is the weighted distance of joint velocities  v weights the velocity difference with respect to d(p i,p j )  d(p i,p j ) is the weighted difference of joint orientations.  p: global translational positions of the figure  q: are the orientations of joint k expressed as quaternions.

6 Alivn/GAME Lab./CSIE/NDHUAn Evaluation of a Cost Metric for Selecting Transitions between Motion Segments 6 Optimization Input  16 different segments of captured motion.  Each is several seconds long. Select  16 good transitions  26 bad transitions  By a single expert and examined by two other viewers for approval.

7 Alivn/GAME Lab./CSIE/NDHUAn Evaluation of a Cost Metric for Selecting Transitions between Motion Segments 7 Optimization (cont.) Method  Constrained least-squares minimization. The weights were non-negative and symmetric.  Solved using an active set method.

8 Alivn/GAME Lab./CSIE/NDHUAn Evaluation of a Cost Metric for Selecting Transitions between Motion Segments 8 Optimization (cont.) Result  The velocity term makes little effective difference in the cost.  Most of the weights were unimportant.

9 Alivn/GAME Lab./CSIE/NDHUAn Evaluation of a Cost Metric for Selecting Transitions between Motion Segments 9 Evaluation Cross-Validation  Robust  Generalize User Study  Quality Determination

10 Alivn/GAME Lab./CSIE/NDHUAn Evaluation of a Cost Metric for Selecting Transitions between Motion Segments 10 Cross-Validation Optimized with one set of training data deleted. The resulting weights are then used to compute the optimal value of a transition for the deleted data set. The average deviation was less than one frame. The median deviation was zero frames. Compute the same optimal transition in 41 of 42 cases.

11 Alivn/GAME Lab./CSIE/NDHUAn Evaluation of a Cost Metric for Selecting Transitions between Motion Segments 11 User Study 26 non-expert adult. Show an example using the same model. Rate how noticeable and natural the discontinuities were, both individually and in comparison to another motion. Show two six-second motions; the transition points are at t=3s. No interpolation or smoothing was done.

12 Alivn/GAME Lab./CSIE/NDHUAn Evaluation of a Cost Metric for Selecting Transitions between Motion Segments 12 User Study (cont.) Influential Factors  Blending Linear Interpolation Ease-in Ease-out  IK routine For each transition the global position and orientation of the character was matched.

13 Alivn/GAME Lab./CSIE/NDHUAn Evaluation of a Cost Metric for Selecting Transitions between Motion Segments 13 User Study (cont.) Divide the viewers into two equal-sized groups. After viewing each sequence, ask 6 questions using 5-points Likert scale. The result were analyzed using two-way between-subjects analyses of variance.

14 Alivn/GAME Lab./CSIE/NDHUAn Evaluation of a Cost Metric for Selecting Transitions between Motion Segments 14 Conclusion & Future Works Not contain highly dynamic motions. Limit to 16 different sequences. Use only one performer. Not contain backward motions. Velocity is not significant. Examine other cost metrics. The effects of transition mechanisms. How people perceive the transitions.

15 Alivn/GAME Lab./CSIE/NDHUAn Evaluation of a Cost Metric for Selecting Transitions between Motion Segments 15 Evaluation Form 論文簡報部份  完整性介紹 (3)  系統性介紹 (4)  表達能力 (3)  投影片製作 (3) 論文審閱部分  瞭解論文內容 (3)  結果正確性與完整性 (4)  原創性與重要性 (4)  讀後啟發與應用: The resulting weights are very useful for our research. Besides, Cross-Validation is a method to make our experimental results more robust; User-Study can show that our experimental results are better than others.


Download ppt "An Evaluation of a Cost Metric for Selecting Transitions between Motion Segments EG SCA ’03 Speaker: Alvin Date: 4 October 2004."

Similar presentations


Ads by Google