Results 1 - 10 of 43,077

Table 1: Results of Pose Estimation Poses

in Exploiting Self-Adaptive Posture-based Focus Estimation for Lecture Video Editing ABSTRACT
by Feng Wang
"... In PAGE 4: ... Some errors occur when the faces are occluded or projected by the slide images of dark color from the LCD projec- tor. Table1 shows the results of pose estimation. We compare three approaches described in Section 4 and 5: di- rect estimation, estimation with flxed probability, and with self-adaptive probability.... ..."

Table 2: Result of Hand Poses Recognition

in 3-D Drawing System via Hand Motion Recgnition from Two Cameras
by Kiyofumi Abe, Hideo Saito, Shinji Ozawa

Table 1: Pose estimation accuracy

in Synthetic Aperture Radar Automatic Target Recognition with Three Strategies of Learning and Representation
by Qun Zhao, Jose C. Principe, Victor Brennan, Dongxin Xu, Zheng Wang
"... In PAGE 6: ... We con- ducted extensive tests for a one-degree of freedom pose estimation of military vehicles using the MSTAR Database 47,48. Results are summarized in Table1 . We conclude that when the pose esti- mator was trained with views of 2 vehicles (T72 and BMP2) at 3.... ..."

Table 1:Results of Pose Estimation

in Address: Same as above.
by D. Levine, Ajit Rajwade, Mcconnell Engineering Building
"... In PAGE 4: ...List of Tables Table1 :Results of Pose Estimation15 Table 2:Recognition rates with ICP, ICP-variant and LMICP23 Table 3:Recognition rates with ICP and the ICP-Variants after applying the feature-based method as an initial step 23 Table 4:Recognition rates with ICP and ICP Variant after applying SVR as the initial step... In PAGE 39: ....8 degrees2.58 degrees3.2 degrees2.72 degrees Table1 :Results of Pose Estimation Face Recognition Rates (Percentage) Angle of rotation of probes (Around Y-axis)ICPICP Variant (this paper) LMICP +/- 10 degrees83.... ..."

Table 2. The error of pose estimation

in 6 DOF pose estimation of polyhedral objects based on analysis of geometric features in x-ray images
by Jae Wan Kim, Young Jun Roh, Hyung Suck Cho
"... In PAGE 10: ... SIMULATION RESULTS G67b We conducted a series of simulations on three different sample objects, which are a cube, a pyramid and an octahedral object of which x-ray images are shown in figure 11. Table2 represents the simulation results for these three objects, where the line features are mainly used in the estimation and corner points are partially used. The simulations are conducted on 55 different poses for each object, and the averaged errors of the results are listed in the table.... ..."

Table 1. Confusion matrix for the articulation estimate (left/upper part for articulations heading right, right/lower part for the mirrored articulations). Rows correspond to the annotations, columns to the estimates.

in Multi-Aspect Detection of Articulated Objects
by unknown authors
"... In PAGE 6: ... There an estimation perfor- mance of 55% is achieved for the same number of clusters. Table1 shows the confusion matrix for the articulation estimates with the 4D-ISM. The upper left part of the ma- trix corresponds to articulations heading right in the order depicted in Figure 4.... ..."

Table 3. Accuracy of the pose estimates for

in Estimating Anthropometry and Pose fr om a Single Image
by unknown authors

Table 1: Classi cation and pose estimation of four at-shaped NIST CAD machine parts.

in Automated Estimation of Class and Pose of Machined Parts
by In Factory
"... In PAGE 2: ...raining set, and the intermediate aspect views (5o, 15o,...355o) as the test set. For purposes of comparison, the class cation and pose estimates obtained using standard KL and FK features are also included. We rst constructed FSTs for each part using linear MRDFs for discrimination only (k=0) (see the left hand side of Table1 ); Classi cation PC is excellent with our new MRDFs performing better than standard FK and KL features (FK features are expected to be better than KL features for discrimination), PC = 97.... In PAGE 2: ... 2a, many test set aspect views had 180o pose estimate errors, due to the symmetry of the object. The data on the right hand side of Table1 shows that our features with k=0.5 gave both good classi cation and pose estimates and that nonlinear features give the best pose estimates.... ..."

Table 1 shows the results of a quantitative comparison. In a sequence of 450 frames where a moving hand changed its state four times, the result of automatic hand tracking was compared with a manually determined ground truth. While the position of the hand is correctly determined in most frames without using colour prior, the pose is often misclassified. After adding the prior on skin colour, we see a substantial improvement in both position and pose.

in Hand-gesture recognition using multi-scale colour features hierarchical features and particle filtering
by Lars Bretzner, Ivan Laptev, Tony Lindeberg 2002
"... In PAGE 4: ... Table1 : Results of a quantitative evaluation of the performance of the hand tracker in a sequence with 450 frames, with and without a prior on skin colour. The errors in the pose estimate that remain occur spuri- ously, and in the prototype system described next, they are reduced by temporal filtering, at the cost of slower dynam- ics when capturing state changes.... ..."
Cited by 8

Table 1 Attributes of the hand-pose dataset extracted in Ref. [16]

in Attribute bagging: improving accuracy of classifier ensembles by . . .
by Robert Bryll, Ricardo Gutierrez-Osuna , Francis Quek 2001
"... In PAGE 3: ... All attributes have been discretized into seven possible values. The attributes, which are summarized in Table1 , were computed for the hand area after the image was segmented based on color information. Fig.... ..."
Results 1 - 10 of 43,077