Narrow Results

To generate a 3D computer model of a free-form object, multiple range images (point clouds) covering its entire surface are acquired from different viewpoints. These views are then aligned in a common coordinate basis by minimizing the distance error between their corresponding points. Establishing correspondences automatically is an inherently challenging problem due to the lack of any type of information other than the geometrical information extracted from the point clouds. Existing "automatic" correspondence techniques achieve automatism at the expense of other important specifications namely, applicability to free-form objects, accuracy, efficiency, robustness to resolution and surface sampling, robustness to overlap, robustness to noise and finally their applicability to simultaneous multiview correspondence. There is also a lack of a review paper that describes and critically analyzes these techniques. In this paper, we present such an extensive review and carry out the analysis of each technique according to the above listed indispensable criteria. Our analysis shows that none of these techniques fully meets these criteria and that there is still a need for the development of practical automatic correspondence algorithms.

Two novel methods for fully unsupervised human action retrieval using 3D mesh sequences are presented. The first achieves high accuracy but is suitable for sequences consisting of clean meshes, such as artificial sequences or highly post-processed real sequences, while the second one is robust and suitable for noisy meshes, such as those that often result from unprocessed scanning or 3D surface reconstruction errors. The first method uses a spatio-temporal descriptor based on the trajectories of 6 salient points of the human body (i.e. the centroid, the top of the head and the ends of the two upper and two lower limbs) from which a set of kinematic features are extracted. The resulting features are transformed using the wavelet transformation in different scales and a set of statistics are used to obtain the descriptor. An important characteristic of this descriptor is that its length is constant independent of the number of frames in the sequence. The second descriptor consists of two complementary sub-descriptors, one based on the trajectory of the centroid of the human body across frames and the other based on the Hybrid static shape descriptor adapted for mesh sequences. The robustness of the second descriptor derives from the robustness involved in extracting the centroid and the Hybrid sub-descriptors. Performance figures on publicly available real and artificial datasets demonstrate our accuracy and robustness claims and in most cases the results outperform the state-of-the-art.