They provide an extensible framework to reconstruct shapes.
Second row: reconstructed shape and texture global AAM.
The proposed reflection model operates on an image dataset to output the reconstructed shape.
As can be noted in Figure 19b,c, these light spots lead to some peaks in the reconstructed shape.
Those results fully meet our expectations and conform to the characteristics of PLDA, which provide optimally and more meaningfully reconstructed shape bases than those given by PPCA.
This can allow (i) to correct the reconstructed shape in the features points neighbourhoods and (ii) to extract the absolute size and distance of the object under evaluation.
In parallel to the feature-based matching algorithm, the processor subsystem performs the novel fast SfS algorithm, presented in the previous section, and provides in output the reconstructed shape of the actual observed object portion.
Moreover, in this type of missions, high performances are required to maximize the throughput of processed frames that enables an increase of the extracted information and, consequently, the accuracy of the reconstructed shape models.
For example, Aanaes et al. [14] impose the prior knowledge that the reconstructed shape does not vary much from frame to frame while Del Bue et al. [15] impose the constraint that some of the points on the object are rigid.
Based on the simulation, three different geometric features, i.e. maximum deformation angle, maximum deformation depth and width of displacement constraint of the reconstructed shape of the deformed body are extracted.
Nevertheless, the resulting shape provided by the Fast-SfS algorithm follows the profile of the real object and the peaks can be efficiently corrected by merging the reconstructed shape with the information gathered from the stereo-vision algorithm (as discussed in Section 'Related works').
