Dense Variational Reconstruction of Non-rigid Surfaces from Monocular Video

By Ravi Garg, Anastasios Roussos, Lourdes de Agapito. 2013

244 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

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

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Category-specific object reconstruction from a single image (2014, background): “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): “Recent results in nonrigid SfM such as [92], [97] are less restrictive but only work in small scenarios.”
Unsupervised CNN for Single View Depth Estimation: Geometry to the Rescue (2016, methodology): “Our autoencoder’s reconstruction loss can be seen as a major generalization of the multiframe optic ﬂow methods like [9,8].”
Non-Rigid Neural Radiance Fields: Reconstruction and Novel View Synthesis of a Dynamic Scene From Monocular Video (2020, background): “Dense non-rigid structure from motion requires dense point tracks over input images, which are then factorized into camera poses and non-rigid 3D states per view [17, 32, 68].”
Lifting AutoEncoders: Unsupervised Learning of a Fully-Disentangled 3D Morphable Model Using Deep Non-Rigid Structure From Motion (2019, methodology): “NRSfM, developed originally to establish a 3D model of a deformable object by observing its motion [4], was developed to solve increasingly accurately the underlying mathematical optimization problems [43, 30, 1, 9], extending to dense reconstruction [11], lifting object cat-”
Simultaneous Optical Flow and Intensity Estimation from an Event Camera (2016, methodology): “We following here the minimisation strategy for a biconvex function in [9] by minimising this term by alternating between the estimation for L and u.”
VNect (2017, unknown): “Structure-from-motion techniques exploit motion cues in a batch of frames [Garg et al. 2013],”
Total Moving Face Reconstruction (2014, background): “Non-rigid structure from motion methods enable reconstruction from a single video by creating a linear basis for the non rigid motion that appears in the particular video; correspondence between the frames is typically given [14,13,20] or estimated via optical flow [23].”
VolumeDeform: Real-Time Volumetric Non-rigid Reconstruction (2016, background): “Non-rigid structure-from-motion methods can capture dense deforming geometry from monocular RGB video [41]; however, results are very coarse and reconstruction is far from real-time.”
Using Locally Corresponding CAD Models for Dense 3D Reconstructions from a Single Image (2017, methodology): “For the problem of dense 3D shape reconstruction, notable examples include [7] and [13] both of which extended the above methods by employing optical flow in order to establish denser correspondences.”
Learning Category-Specific Deformable 3D Models for Object Reconstruction (2017, background): “Originally proposed for recovering shape and deformations from video [40], [41], [42], [43], NRSfM is a natural choice for camera estimation”
Mnemonic Descent Method: A Recurrent Process Applied for End-to-End Face Alignment (2016, background): “Non-linear least squares optimisation problems often arise in computer vision, including but not limited to Structure-from-Motion [9, 23], rigid and deformable image alignment [33, 35], optical flow estimation [33, 66, 46, 45], CNN RNN”
Video Pop-up: Monocular 3D Reconstruction of Dynamic Scenes (2014, background): “On the other hand, although NRSfM algorithms can reconstruct a single pre-segmented deformable surface moving in front of a camera [14], [39], they require segmentation of the scene into background and foreground objects.”