Academia.eduAcademia.edu

Figure 12 – uploaded by Anusha Jayasiri

See full PDFdownloadDownload figure
Resulting force feedback of the SPIDAR for the image equence calculated using equation (7) is shown in Fig. 13. Notably, resulting force is within the sensible region of the SPIDAR-G and hence user can smoothly feel the movement of he objects in the video. Fig. 13Feedback Force generated from SPIDAR forthe bouncing ball using a non linear function We can compare the resulting feedback force using two methods: a gain controller and non-linear function with respect to the position and velocity changes by using Fig. 11 and Fig. 13. It is clear that from Fig. 13, nonlinear function method outperforms the gain controller method,since it increases the feedback force for low velocities than the gain controller method and decreases the feedback force for high velocities

Figure 13 Resulting force feedback of the SPIDAR for the image equence calculated using equation (7) is shown in Fig. 13. Notably, resulting force is within the sensible region of the SPIDAR-G and hence user can smoothly feel the movement of he objects in the video. Fig. 13Feedback Force generated from SPIDAR forthe bouncing ball using a non linear function We can compare the resulting feedback force using two methods: a gain controller and non-linear function with respect to the position and velocity changes by using Fig. 11 and Fig. 13. It is clear that from Fig. 13, nonlinear function method outperforms the gain controller method,since it increases the feedback force for low velocities than the gain controller method and decreases the feedback force for high velocities

Related Figures (14)

Each string is connected to a motor and an encoder at one end and tothe grip at the other end. The feedback force is determined by the tension of each string generated by the motor, which is transformedto the users hand through the grip. By connecting this device to a personal computer, it provides a high definition force feedback sensation to the user [1][14].
Each string is connected to a motor and an encoder at one end and tothe grip at the other end. The feedback force is determined by the tension of each string generated by the motor, which is transformedto the users hand through the grip. By connecting this device to a personal computer, it provides a high definition force feedback sensation to the user [1][14].
Fig.2Haptic interaction scenario As a result, the viewercan feel the movement of the objects in the video or the forces related to the movement into his/her hand rather than just seeing it as illustrated in Fig. 2.
Fig.2Haptic interaction scenario As a result, the viewercan feel the movement of the objects in the video or the forces related to the movement into his/her hand rather than just seeing it as illustrated in Fig. 2.
The proposed approach for this research can be illustratec using a block diagram as in Fig. 3.
The proposed approach for this research can be illustratec using a block diagram as in Fig. 3.
Feature points selection is an important task of any computer vision and image processing application. Since feature selection is the starting point of many computer vision
Feature points selection is an important task of any computer vision and image processing application. Since feature selection is the starting point of many computer vision
Lucas-Kanade method 1s an example tor this type of optical flow [4] [18].We used Pyramid Lucas-Kanade Algorithm, which is a pyramidal implementation of the Lucas-Kanade feature tracker [5]. At first in this technique, it solves optical flow at the top layer of the pyramid and then use the resulting motion estimates as the starting point for the next layer down. It continues going down the pyramid in this manner until it reaches the lowest level. Therefore by using this method, it can track faster and longer motions [4]. This has less computation and therefore it could be easily adapted for real time applications [2].
Lucas-Kanade method 1s an example tor this type of optical flow [4] [18].We used Pyramid Lucas-Kanade Algorithm, which is a pyramidal implementation of the Lucas-Kanade feature tracker [5]. At first in this technique, it solves optical flow at the top layer of the pyramid and then use the resulting motion estimates as the starting point for the next layer down. It continues going down the pyramid in this manner until it reaches the lowest level. Therefore by using this method, it can track faster and longer motions [4]. This has less computation and therefore it could be easily adapted for real time applications [2].
Fig. 6 Two consecutive frames of an image sequence Bouncing ball is a good example for position, velocity and acceleration change against time because these graphs convey lots of information about speeding up, slowing down, rising or falling of the ball.
Fig. 6 Two consecutive frames of an image sequence Bouncing ball is a good example for position, velocity and acceleration change against time because these graphs convey lots of information about speeding up, slowing down, rising or falling of the ball.
Fig. 10Performance of the k value We analysed the pattern of k value using the previously mentioned image sequence of the bouncing ball. Fig. 10 shows the results of the changing k value for the image sequence. Resulting force feedback of the SPIDAR for the image
Fig. 10Performance of the k value We analysed the pattern of k value using the previously mentioned image sequence of the bouncing ball. Fig. 10 shows the results of the changing k value for the image sequence. Resulting force feedback of the SPIDAR for the image
The Fig. 7, 8 and 9 complies with the physics of a bouncing ball [9] and hence we can conclude that our proposed method is accurate. Fig. 7Position changes in bouncing ball
The Fig. 7, 8 and 9 complies with the physics of a bouncing ball [9] and hence we can conclude that our proposed method is accurate. Fig. 7Position changes in bouncing ball
Fig. 8 Velocity changes in bouncing ball
Fig. 8 Velocity changes in bouncing ball
D. Haptic Motion Rendering
D. Haptic Motion Rendering
Fig.12 Purpose of the non-linear function Similar to the previous method, the purpose of using a nonlinear function is to maintain the feedback force in the sensible region by decreasing the feedback force for high velocities and increasing the feedback force for low velocities. Fig. 12 illustrate this idea in a graphical form. The International Conference on Advances in ICT for Emerging Regions — ICTer2012
Fig.12 Purpose of the non-linear function Similar to the previous method, the purpose of using a nonlinear function is to maintain the feedback force in the sensible region by decreasing the feedback force for high velocities and increasing the feedback force for low velocities. Fig. 12 illustrate this idea in a graphical form. The International Conference on Advances in ICT for Emerging Regions — ICTer2012
Fig. 11Feedback Force generated from SPIDAR forthe bouncing ball using the gain controller method sequence calculated using equation (3) is shown in Fig 11. Notably, resulting force is within the sensible region of the SPIDAR-G and hence user can smoothly feel the movement of the objects in the video. Il) Method Using a Nonlinear Function
Fig. 11Feedback Force generated from SPIDAR forthe bouncing ball using the gain controller method sequence calculated using equation (3) is shown in Fig 11. Notably, resulting force is within the sensible region of the SPIDAR-G and hence user can smoothly feel the movement of the objects in the video. Il) Method Using a Nonlinear Function
Fig. 15 Distribution of positive responses for using a non-linear function method Fig. 15 shows the distribution of positive responses for the non-linear function method, which was the selected method by most users as the better method out of the two.
Fig. 15 Distribution of positive responses for using a non-linear function method Fig. 15 shows the distribution of positive responses for the non-linear function method, which was the selected method by most users as the better method out of the two.
Fig. 14 Result of responses of users for each method The result of this rating is shown in Fig. 14. This grapl shows how users rate each method based on their feeling a ‘Good * or ‘Bad’. We tested with the involvement of tet numbers of real participants and it is clear thatmore than 60% of users responded that using a non-linear function is bette than using a gain controller.
Fig. 14 Result of responses of users for each method The result of this rating is shown in Fig. 14. This grapl shows how users rate each method based on their feeling a ‘Good * or ‘Bad’. We tested with the involvement of tet numbers of real participants and it is clear thatmore than 60% of users responded that using a non-linear function is bette than using a gain controller.
Related topics:
Haptic InteractionMultimedia Applications
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved