Q. Cai and J. K. Aggarwal. Tracking human motion using multiple cameras. In ICPR96, pages 68--72, 1996.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....think he should have included a underlying explicit model of the human because it could have solved some of the occlusion problem which he must have. It would also help is probable necessary when he wants to expand his algorithm to 3D. Title: Tracking Human Motion Using Multiple Cameras [17] Author(s) Q. Cai and J.K. Aggarwal Location: Computer and Vision Research Center, University of Texas, Austin, USA Year: 1996 Published: International Conference on Pattern Recognition Type: Paper Key words: Multiple human tracking, multiple cameras, human model, double difference method, ....

Q. Cai and J.K Aggarwal. Tracking Human Motion Using Multiple Cameras. In International Conference on Pattern Recognition, 1996.

.... Kameda et al. 85] 1995 Leung and Yang [98] 1995 Pentland [122] 1995 Pentland [123] 1995 Tesei et al. 136] 1996 Azarbayejani et al. 9] 1996 Becker and Pentland [14] 1996 Bobick [19] 1996 Bobick and Davis [21] 1996 Cai and Aggarwal [27] 1996 Gavrila and Davis [55] 1996 Ju [77] 1996 Ju et al. 78] 1996 Kakadiaris and Metaxas [82] 1996 Kameda and Minoh [83] 1996 Luc [46] 1996 Moezzi et al. 109] 1996 Turk [138] 1996 Wren et al. 144] 1997 Aggarwal ....

....parallel to the camera ffl No occlusion allowed ffl Slow and continuous movements ffl The workspace is flat The first three assumptions are very general and they are used in nearly every system. There are, however, a few systems which are devoted specifically to these problems, see e.g. 44] [27], and [114] The fourth assumption is mainly used in HCI applications where the user is facing a screen. The assumption makes it possible to calculate the overall body pose from a set of simple rules. The next assumption is used a lot in gait analysis since it reduces the dimensionally of the ....

Q. Cai and J.K Aggarwal. Tracking Human Motion Using Multiple Cameras. In International Conference on Pattern Recognition, 1996.

....1. Known start pose 10. Subject moves on a flat ground plane 2. Known subject 3. Markers placed on the subject 4. Special coloured clothes 5. Tight fitting clothes The first three assumptions related to movements are very general and used in every system with a few exceptions, see e.g. 36] [19], and [104] The fourth assumption is mainly used in HCI applications, and simplifies the calculation of the overall body pose. The next assumption reduces the dimensionally of the problem from 3D to 2D and is often used in applications such as gait analysis. The sixth assumption concerns ....

Q. Cai and J.K Aggarwal. Tracking Human Motion Using Multiple Cameras. In International Conference on Pattern Recognition, 1996.

....Tracking involves a loop of predicting the appearance of the person in the new image, determining for each pixel the likelihood that it is part of one of the blob models or background model, assigning it to one of the models, and updating the statistical models. See Figure 9. Cai and Aggarwal [11] describe a system with a simplified head trunk model to track humans across multiple cameras. In this work, tracking uses point features derived from the medial axis of the foreground region. Attributes used for tracking are position and velocity of the points, together with the average intensity ....

.... social interactions such as shaking hands and dancing) It is unlikely that this 21 2 D approaches without 2 D approaches with 3 D approaches explicit shape models explicit shape models Baumberg and Hogg [8] Akita [3] Azarbayejani and Pentland [4] Bobick and Wilson [10] Cai and Aggarwal [11] Campbell and Bobick [13] Charayaphan and Marble [16] Chang and Huang [15] Chen and Lee [17] Cootes et al. 18] Geurtz [27] Dorner [21] Darell and Pentland [19] Goddard [28] Downton and Drouet [22] Davis and Shah [20] Guo et al. 30] Gavrila and Davis [25] 26] Franke et al. 23] Herman [34] ....

Q. Cai and J. Aggarwal. Tracking human motion using multiple cameras. In Proc. of International Conference on Pattern Recognition, pages 68--72, Vienna, 1996.

....Tracking involves a loop of predicting the appearance of the person in the new image, determining for each pixel the likelihood that it is part of one of the blob models or background model, assigning it to one of the models, and updating the statistical models. See Fig. 9. Cai and Aggarwal [11] described a system with a simplified head trunk model to track humans across multiple cameras. In this work, tracking uses point features derived from the medial axis of the foreground region. Attributes used for tracking are position and velocity of the points, together with the average ....

.... human moveTABLE 2 A Selection of Previous Work on the Visual Analysis of Human Movement 2 D approaches without 2 D approaches with explicit shape models explicit shape models 3 D approaches Baumberg and Hogg [8] Akita [3] Azarbayejani and Pentland [4] Bobick and Wilson [10] Cai and Aggarwal [11] Campbell and Bobick [13] Charayaphan and Marble [16] Chang and Huang [15] Chen and Lee [17] Cootes et al. 18] Geurtz [27] Dorner [21] Darell and Pentland [19] Goddard [28] Downton and Drouet [22] Davis and Shah [20] Guo et al. 30] Gavrila and Davis [25] 26] Franke et al. 23] Herman [34] ....

Q. Cai and J. Aggarwal, Tracking human motion using multiple cameras, in Proc. of International Conference on Pattern Recognition, Vienna, 1996, pp. 68--72.

.... out that the majority of today s known tracking algorithms mainly evaluate the motion information in the image sequence (see [16] either derived from difference images or from computation of the optical flow ( 8] There are several examples for the successful use of this approach (see e.g. [2, 7, 9, 19]) However, as soon as there is other motion in the image besides the moving object, this approach can lead to severe difficulties. This can be very frequently the case in realistic scenarios, such as e.g. the surveillance of traffic, corridors, shopping malls, gas stations or parking lots. Here, ....

Q. Cai and J. Aggarwal. Tracking Human Motion Using Multiple Cameras. In Proceedings of ICPR, volume 3, pages 68--72, Vienna, 1996.

....of view of one camera, it will appear in the view of another camera in the system. A multiple camera setup also helps to solve the ambiguity of matching when subject images are occluded to each other. Only in very recent years has work on tracking of human motion from multiple perspectives emerged [43, 22, 27, 9]. Compared to the problem of tracking moving humans from a single camera, establishing feature correspondence between images captured at different locations is more challenging because the features are recorded in different spatial coordinates. All features to be tracked must be adjusted to the ....

....feature correspondence between images captured at different locations is more challenging because the features are recorded in different spatial coordinates. All features to be tracked must be adjusted to the same spatial reference before matching is performed. Recent work by Cai and Aggarwal [9] uses multiple points belonging to the medial axis of the human upper body as the feature to track. These points are sparsely sampled and assumed to be independent of each other, which preserves a certain degree of nonrigidity of the human body. Location and average intensity of the feature ....

[Article contains additional citation context not shown here]

Q. Cai and J. K. Aggarwal. Tracking human motion using multiple cameras. In Proc. of Intl. Conf. on Pattern Recognition, pages 68--72, Vienna, Austria, August 1996.

No context found.

Q. Cai and J. K. Aggarwal. Tracking human motion using multiple cameras. In ICPR96, pages 68--72, 1996.

No context found.

J. K. Aggarval and Q. Cai. Tracking human motion using multiple cameras. In Proceedings on Pattern Recognition - ICPR 2000, volume 4, pages 184--187, Barcelona, Spain, September 3-8 2000.

No context found.

Q. Cai and J. K. Aggarval. Tracking human motion using multiple cameras. In Proceedings of the 13 on Pattern Recognition - ICPR 2000, volume 4, pages 84-- 87, Barcelona, Spain, September 3-8 2000.

No context found.

Q. Cai and J. Aggarwal, "Tracking human motion using multiple cameras", Proc. ICPR, Vienna, pp.68-72, 1996.

No context found.

Cai Q and Aggarwal J K, Tracking human motion using multiple cameras, in Proc. of the 13th International Conference on Pattern Recognition (1996) 68-72.

No context found.

Q. Cai and J. K. Aggarwal. Tracking human motion using multiple cameras. In Proc. of the 13th International Conference on Pattern Recognition, pages 68-72, 1996.

No context found.

Q. Cai and J. K. Aggarwal, "Tracking human motion using multiple cameras," in Proc. of the 13th International Conference on Pattern Recognition, pp. 68-72, 1996.

No context found.

Q. Cai and J. Aggarwal. Tracking human motion using multiple cameras, 1996.

No context found.

Q. Cai and J. Aggarwal, "Tracking human motion using multiple cameras", Proc. ICPR, Vienna, pp.68-72, 1996.

No context found.

Q. Cai, J.K. Aggarwal, Tracking human motion using multiple cameras, Proceedings of the 13th International Conference on Pattern recognition, 1996, pp. 68 -- 72.

No context found.

Q. Cai and J. K. Aggarwal, "Tracking human motion using multiple cameras," in Proc. 13th Int. Conf. Pattern Recognition, Los Alamitos, CA, Aug. 1996, pp. 68--72.

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