### Table 1. Comparison of 3-D RMS error between unconstrained and constrained stereo results (n is the number of points).

1996

"... In PAGE 12: ...Itisalsoworth noting that the accuracy of the iterative algorithm is best at smaller numbers of input points, suggesting that it is more stable given a smaller number of input data. Table1 lists the 3-D errors of both constrained and unconstrained (8-point only) methods for the synthetic scenes. It appears from this result that the constrained method yields better results (af- ter median ltering) and more points (a result of reducing the aperture problem).... ..."

Cited by 101

### Table 3. Results for stereo

"... In PAGE 5: ... The cost of labelling a pixel as oc- cluded was fixed at a16 a35a4 in all cases. Table3 shows the per- centage of bad matching pixels obtained as a result of ap- plying our algorithm to the four data sets available. For comparison purpose, the results obtained from other dense stereo algorithms have also been included.... ..."

### Table 2: Results of estimated motion parameters from stereo images 4.7 Conclusions Several algorithms for motion extraction have been described in this chapter. For pure translation and pure rotation, linear methods are still favourable and can give reasonable results. For general cases, the availability of two pairs of stereo images leads to a linear method which can give much more reliable results and which does not su er from the ambiguous scale problem existing for any method using a pair of monocular images.

"... In PAGE 65: ... The translation vector T is then calculated using Equation 73 once the rotation para- meters are available. The simulation has run 100 times with random noise, and the mean values of the calculated motion parameters are shown in Table2 together with their true values. From the table it can be seen that the estimated motion parameters are very close to their true values although the 3D point coordinates are corrupted with noise.... ..."

### Table 2. Stereo control commands.

2005

"... In PAGE 4: ...chose a set of 20 commands that could be used to control an in-car stereo system (see Table2 ). The reason station three is not on the list is because the subset of articulatory features we are currently using cannot distinguish it from next station .... ..."

Cited by 7

### Table 5: The accuracy (%) of the stereo matcher using multiple window stereo algorithm

2000

Cited by 32

### Table 2: Stereo correspondence results

"... In PAGE 4: ... Any points for which the least squares matching solution di- verges, or does not converge fast enough (a threshold of 20 iterations is currently used), are also invalidated, and removed from the set of correspondences. Results from the stereo correspondence, using the above matching strategy, are shown in Table2 . A graphical depic- tion of the nal point correspondences for the image pair is depicted in Figure 4.... ..."

### Table 5.1 Tested Stereo Video Sequences Stereo Video

2006

### Table 4: The accuracy (%) of the stereo matcher

2000

"... In PAGE 19: ... As one can see from Table 3 the Cauchy distribution has the best t to the measured distribution. Therefore, one expects the accuracy to be the greatest when using Lc ( Table4 ). In all cases (Figure 13) the results obtained with L2 are the worst.... ..."

Cited by 32

### Table 3. The accuracy of the stereo matcher

1998

Cited by 4

### Table 3. The accuracy of the stereo matcher

1998

Cited by 4