Results 1 - 10 of 30,953

Table 3. Comparison of data size between multi-view frames and LDI frame [kBytes]

in A framework for multi-view video coding using layered depth image
by Seung-uk Yoon, Eun-kyung Lee, Sung-yeol Kim, Yo-sung Ho 2005
"... In PAGE 10: ... There are no holes in the first layer but holes are increased as the number of layers increase. Table3 lists the data size between multi-view frames of the test sequence and the generated LDI frame [17]. In Table 3, the sum of frames implies the summation of the data size of eight color and eight depth images.... In PAGE 10: ... Table 3 lists the data size between multi-view frames of the test sequence and the generated LDI frame [17]. In Table3 , the sum of frames implies the summation of the data size of eight color and eight depth images. It is clear that the data size is reduced by converting multi-view video to LDI frames even without any encoding process.... ..."
Cited by 2

Table 3. Test results of the multi-view face de- tector and pose estimator from a total of over 1000 images from the test-sequence set.

in Multi-View Face Detection and Pose Estimation Using A Composite Support Vector Machine across the View Sphere
by Jeffrey Ng, Shaogang Gong 1999
"... In PAGE 5: ... Figures 6, 7 and 8 show example frames from dif- ferent test sequences in which novel faces were detected in multi-views and tracked across views with their pose esti- mated simultaneously. Table3 provides a summary of the results for the test sequences and it can be noticed that the detection rate and the average pose estimation error do not vary significantly between the sequences of known subjects and those of unknown subjects, namely sequences A and B against C and D. Sequence A Sequence C Sequence D Figure 6.... ..."
Cited by 19

Table 1. The division of the view sphere for learning multi-view SVMs.

in Multi-View Face Detection and Pose Estimation Using A Composite Support Vector Machine across the View Sphere
by Jeffrey Ng, Shaogang Gong 1999
"... In PAGE 3: ... Such nega- tive samples are then used to train a multi-view face model based on a set of local component SVMs along the view sphere. Given the face distribution in PES as shown in Figure 3, the view sphere can be divided into smaller, more localised yaw segments as in Table1 . The observed asymmetry of the view sphere distribution and the greater complexity of the left portion are reflected into the selection of smaller... ..."
Cited by 19

Table 1. Cost for a three person MultiView site.

in MultiView: Spatially Faithful Group Video Conferencing
by John Canny 2005
Cited by 5

TABLE I MULTI-VIEW GEOMETRIC CONSTRAINTS FOR MOTION DETECTION

in Detecting motion regions in presence of strong parallax from a moving camera by multi-view geometric constraints
by Chang Yuan, Gérard Medioni, Jinman Kang, Isaac Cohen
Cited by 2

Table 2 Parameters used to train the multi-view face detectors

in Support vector machine based multi-view face detection and recognition
by Yongmin Li , Shaogang Gong , Jamie Sherrah , Heather Liddell

Table 1: Comparisons of our results with other multi-view detectors. Each column shows the detec- tion rates for a given average number of false positives per image (these rates correspond to those for which other authors have reported results). Results for real-time detectors are shown in bold. Note that ours is the only single detector that can be tested on all data sets simultaneously.

in Synergistic Face Detection and Pose Estimationwith Energy-Based Models Margarita OsadchyNEC Labs America
by unknown authors
"... In PAGE 7: ... Figure 4 shows a few results on various poses. Table1 shows our detection rates compared against other systems for which results were given on these data sets. The table shows that our results on the TILTED and PROFILE sets are similar to those of the two Jones amp; Viola detectors, and even approach those of the Rowley et al and Schneiderman amp; Kanade non- real-time detectors.... ..."

Table 1: Comparisons of our results with other multi-view detectors. Each column shows the detec- tion rates for a given average number of false positives per image (these rates correspond to those for which other authors have reported results). Results for real-time detectors are shown in bold. Note that ours is the only single detector that can be tested on all data sets simultaneously.

in Abstract
by Margarita Osadchy, Matthew L. Miller, Yann Le Cun
"... In PAGE 7: ... Figure 4 shows a few results on various poses. Table1 shows our detection rates compared against other systems for which results were given on these data sets. The table shows that our results on the TILTED and PROFILE sets are similar to those of the two Jones amp; Viola detectors, and even approach those of the Rowley et al and Schneiderman amp; Kanade non- real-time detectors.... ..."

Table 1: Comparisons of our results with other multi-view detectors. Each column shows the detec- tion rates for a given average number of false positives per image (these rates correspond to those for which other authors have reported results). Results for real-time detectors are shown in bold. Note that ours is the only single detector that can be tested on all data sets simultaneously.

in Abstract
by Margarita Osadchy, Matthew L. Miller, Yann Le Cun
"... In PAGE 7: ... Figure 4 shows a few results on various poses. Table1 shows our detection rates compared against other systems for which results were given on these data sets. The table shows that our results on the TILTED and PROFILE sets are similar to those of the two Jones amp; Viola detectors, and even approach those of the Rowley et al and Schneiderman amp; Kanade non- real-time detectors.... ..."

Table 2: Fixed cost for building a MultiView site. These are the materials needed to build the screen.

in MultiView: Spatially Faithful Group Video Conferencing
by David Nguyen 2005
"... In PAGE 27: ...4 Cost One of the benefits of MultiView is its relatively low cost and potentially high gain. The fixed cost is building the MultiView screen, costing around $700 ( Table2 ). The variable costs include the cameras and the projectors (Table 3).... ..."
Cited by 5
Results 1 - 10 of 30,953