### Table 1: Variable and constant definitions for the robust point matching algorithm

"... In PAGE 9: ..., 1991), we have not done so here. Variable and constant definitions can be found in Table1 . The pseudocode describes the RPM algorithm in minute detail.... ..."

### Table 1. Evaluation of algorithm performance using Stirmark Attack Proposed Method Robust Template Matching [8] Digimarc Suresign Unige

2001

"... In PAGE 3: ...00 MHz. This is sufficiently fast for commercial applications. We note here that if the algorithm does not search for geometric distor- tions of the test image then the watermark detection is performed in real time. The results are summarized in Table1 . The number assigned at each attack is the average correct watermark detection of all ex- periments conducted for this attack.... In PAGE 4: ... It is worth noting that the proposed watermarking method can deal with the random geometric distortions applied using Stirmark, since in all cases the watermark is correctly detected. The performance of other commercial products and watermark- ing methods proposed in the literature against the Stirmark bench- marking tests is drawn in Table1 . It can be observed that the pro- posed method attains the best average performance.... ..."

Cited by 5

### Table 1: Theoretical evaluation of robust and non-robust poli- cies on various models, which match empirical evaluation. Normal Pessimistic Optimistic

2005

Cited by 6

### Table 1: Theoretical evaluation of robust and non-robust poli- cies on various models, which match empirical evaluation. Normal Pessimistic Optimistic

2005

Cited by 6

### Table 1: Theoretical evaluation of robust and non-robust poli- cies on various models, which match empirical evaluation. Normal Pessimistic Optimistic

2005

Cited by 6

### Table 2: Construction of 3D Models (H-2,3,4,5,6,7) us- ing image pairs: SM=matches after match strength com- putation, RM=matches after relaxation strategy, EM=robust matches after applying epipolar constraint, IE=Initial repro- jection error, FE= nal reprojection error after bundle ad- justment.

"... In PAGE 7: ...2 (see Figure 12). Statistics are summarized in Table2 . Al- though the initial reprojection error was very high for H5- model because the two images used in the 3D reconstruc- tion differed by a very small rotation, which is not very good for the working of the 3D reconstruction algorithm using epipolar constraint.... ..."

### Table 3: Scale-sensitive image and model based regis- tration for six levels of hierarchy: I=images, FP=feature points, s=scale ratio, RM=matches after relaxation strat- egy, EM=robust matches after applying epipolar constraint, S=scale for 3D registration, E=Least squares registration er- ror

### TABLE IV Equal error rates achieved by elastic graph matching (EGM), morphological elastic graph matching (MDLA) and optimized robust correlation (ORC) when they are applied to MATRANORTEL database. Conditions Number of claims EER (%)

### Table 8 Results of noise robustness experiments

"... In PAGE 4: ...2 (lexicon based on 7 matches, language model trained on 12 matches). Table8 shows the best results for the German YugNet match. Table 8 Results of noise robustness experiments ... ..."

### Table 1. Robust estimation of the epipolar geometry from a set of matches con- taining outliers using RANSAC (POK indicates the probability that the epipolar geometry has been correctly estimated).

1998

"... In PAGE 6: ...Table1... In PAGE 7: ...ion 3.1. Then the matches which correspond to already reconstructed points are used to compute the projection matrix Pk. This is done using a robust pro- cedure similar to the one laid out in Table1 . In this case a minimal sample of 6 matches is needed to compute Pk.... ..."

Cited by 14