Nonrigid Structure-from-Motion: Estimating Shape and Motion with Hierarchical Priors

By Lorenzo Torresani, Aaron Hertzmann, C. Bregler. 2008

501 citations recorded by the source corpus.

Subfields: Computational Mechanics, Computer Vision and Pattern Recognition, 3D Vision & Pose

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Computer Vision and Pattern Recognition

In 3D Vision & Pose

Citations in the corpus

Category-specific object reconstruction from a single image (2014, methodology): “Originally proposed for recovering shape and deformations from video [6, 33, 16, 10], NRSfM is a natural choice for viewpoint estimation from sparse correspondences as intra-class variation may become a confounding factor if not modeled explicitly.”
Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age (2016, neutral): “hape from non-rigid objects but with restrictive applicability, for example Pentland et al. [222, 223] require prior knowledge of some mechanical properties of the objects, Bregler et al. in [31] and [274] requires a restricted deformation of the object and show examples in human faces. Recent resul”
Model evolution: An incremental approach to non-rigid structure from motion (2010, methodology): “These ambiguities have been identified in the study of NRSFM using weak orthographic projection [14, 11].”
Neural volumes (2019, background): “Analogous to non-rigid structure from motion [Torresani et al. 2008], this architecture can leverage a scene’s regularity across time to improve viewpoint generalization further.”
Photo Sequencing (2012, methodology): “A different class of methods use factorization to deal with dynamic scenes [17].”
Learning Category-Specific Deformable 3D Models for Object Reconstruction (2017, methodology): “Formally, our adaptation of the NRSfM algorithm in [42]”
Video Pop-up: Monocular 3D Reconstruction of Dynamic Scenes (2014, background): “Non-rigid structure from motion (NRSfM) approaches reconstruct 3D models of non-rigid objects from monocular video, typically by fitting a global low-rank shape model [37], [26] to 2D tracks.”
Dense Variational Reconstruction of Non-rigid Surfaces from Monocular Video (2013, methodology): “Most approaches have required the addition of extra priors on the shape or camera matrices to resolve the ambiguities such as: temporal smoothness [4, 32], near rigidity (rigid component explains most of the motion) [4, 32], smooth time trajectories [2, 16, 21], basis priors [4] or use of a pre-defi”
A Simple Prior-Free Method for Non-rigid Structure-from-Motion Factorization (2012, methodology): “’s EM-PPCA [25]; (3) Metric projection [21]; (4) Trajectory basis method [3]; and (5) Column space fitting (CSF) [14].”
Finding nemo: Deformable object class modelling using curve matching (2010, background): “Related research comprises several strands: non-rigid structure from motion, rigid 3D reconstruction from contours, and the modelling of shape classes, e.g. using active shape models.”
HyperNeRF (2021, unknown): “Using only a monocular video stream is convenient and inexpensive, but also introduces significant ambiguity which must be ameliorated through the use of factorization [Bregler et al. 2000] or regularization [Torresani et al. 2008].”
Content-preserving warps for 3D video stabilization (2009, methodology): “2002], non-rigid 3D tracking [Torresani et al. 2008], and video hole-filling [Wexler et al.”
3D Constrained Local Model for rigid and non-rigid facial tracking (2012, methodology): “[23], which was created using non-rigid structure from motion [24] approach on the Multi-PIE [17] dataset.”
Inferring 3D Shapes and Deformations from Single Views (2010, background): “’s [21] have attempted to recover non-rigid 3D object shape as in our work but only up to sparse reconstruction using 2D point tracks.”
Factorization for non-rigid and articulated structure using metric projections (2009, methodology): “We show successful results of our algorithms on synthetic and real sequences of both deformable and articulated data.”