### Table 7: Statistics of the pairwise differences between the four scans of the green channel (in grey values).

"... In PAGE 14: ...alues are ca. 1.5 grey values less than the values in Table 6 for similar mean grey value (0.51D). Statistics of the pair- wise differences between the four scans (including all tiles) were also computed. Table7 shows the results for the green channel. The RMS includes the sensor noise and its temporal variation as well as illumination variations.... ..."

### Table 2. Results of the family recognition scan and the corresponding average pairwise alignment quality of the methods

2007

Cited by 1

### Table 3. Behavioral data

1997

"... In PAGE 3: ... Addition- ally, pairwise contrasts between the conditions differentiated by each latent variable were performed separately in each group. RESULTS Behavioral data Behavioral results are shown in Table3 . Recognition was higher than recall, and differences between the two age groups did not reach statistical significance.... ..."

Cited by 16

### Table 1: Evaluation of the performance of the algorithm. A set of pairwise registrations is shown. Each row represents one registered pair of scans. The second column displays the number of line pairs. Column Pre shows the % (over all possible pairs) of line pairs that need to be considered after the preprocessing step of the algorithm. Column S2 shows the % (over all possible combinations) and total number of pairs that reach STAGE 2 and column S3 the same number for STAGE 3, the most expensive stage (in S3 the reduction is computed over all possible pairs of matches ((l1, r1) and (l2, r2))). The efficiency of our algorithm is due to the great reduction of the pairs that need to be considered in this stage. Column M presents the number of matched pairs that the algorithm establishes. The running Time t of the algorithm (in secs) is shown for every pair (2GHz Intel machine). Finally, the pairwise registration Error Err is displayed. This Error is the average distance between matched planar region between the two scans. The error ranges from 1.36mm to 14.96mm for the first data set and from 5.34mm to 56.08mm for the second. Note that the application of the well-known Iterative Closest Point algorithm will improve the results even further.

2003

"... In PAGE 5: ...hown in Fig. 5. Note the registration accuracy. Table1 pro- vides an extensive evaluation of the efficiency and accuracy of our algorithm. The efficiency of the algorithm is demon- strated by the percentage of line pairs that survive after pre- processing, and reach STAGE 2, and STAGE 3 of the algo- rithm.... In PAGE 6: ...details on the performance of pairwise registrations see Table1 .... ..."

Cited by 15