Closed-Form Solutions for Physically Based Shape Modeling and Recognition
IEEE Transactions on Pattern Analysis and Machine Intelligence
A morphable model for the synthesis of 3D faces
Proceedings of the 26th annual conference on Computer graphics and interactive techniques
ECCV '98 Proceedings of the 5th European Conference on Computer Vision-Volume II - Volume II
Uncalibrated Perspective Reconstruction of Deformable Structures
ICCV '05 Proceedings of the Tenth IEEE International Conference on Computer Vision - Volume 2
Nonrigid Structure-from-Motion: Estimating Shape and Motion with Hierarchical Priors
IEEE Transactions on Pattern Analysis and Machine Intelligence
Piecewise quadratic reconstruction of non-rigid surfaces from monocular sequences
ECCV'10 Proceedings of the 11th European conference on Computer vision: Part IV
Monocular template-based reconstruction of smooth and inextensible surfaces
ACCV'10 Proceedings of the 10th Asian conference on Computer vision - Volume Part III
Monocular Template-based Reconstruction of Inextensible Surfaces
International Journal of Computer Vision
Monocular template-based tracking of inextensible deformable surfaces under L2-norm
ACCV'09 Proceedings of the 9th Asian conference on Computer Vision - Volume Part II
Distributed message passing for large scale graphical models
CVPR '11 Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition
Energy based multiple model fitting for non-rigid structure from motion
CVPR '11 Proceedings of the 2011 IEEE Conference on Computer Vision and Pattern Recognition
Monocular 3D Reconstruction of Locally Textured Surfaces
IEEE Transactions on Pattern Analysis and Machine Intelligence
Diverse M-best solutions in markov random fields
ECCV'12 Proceedings of the 12th European conference on Computer Vision - Volume Part V
| Hi-index | 0.00 |
Existing approaches to non-rigid 3D reconstruction either are specifically designed for feature point correspondences, or require a good shape initialization to exploit more complex image likelihoods. In this paper, we formulate reconstruction as inference in a graphical model, where the variables encode the rotations and translations of the facets of a surface mesh. This lets us exploit complex likelihoods even in the absence of a good initialization. In contrast to existing approaches that set the weights of the likelihood terms manually, our formulation allows us to learn them from as few as a single training example. To improve efficiency, we combine our structured prediction formalism with a gradient-based scheme. Our experiments show that our approach yields tremendous improvement over state-of-the-art gradient-based methods.