• Corpus ID: 244102885

Temporally-Consistent Surface Reconstruction using Metrically-Consistent Atlases

  title={Temporally-Consistent Surface Reconstruction using Metrically-Consistent Atlases},
  author={Jan Bednar{\'i}k and Noam Aigerman and Vladimir G. Kim and Siddhartha Chaudhuri and Shaifali Parashar and Mathieu Salzmann and P. Fua},
We propose a method for unsupervised reconstruction of a temporally-consistent sequence of surfaces from a sequence of time-evolving point clouds. It yields dense and semantically meaningful correspondences between frames. We represent the reconstructed surfaces as atlases computed by a neural network, which enables us to establish correspondences between frames. The key to making these correspondences semantically meaningful is to guarantee that the metric tensors computed at corresponding… 


Temporally Coherent Segmentation of 3D Reconstructions
This paper addresses the problem of segmenting consistently an evolving 3D scene reconstructed individually at different time-frames, and works with mesh-based representations, though the method can be extended to other 3D representations as well.
Unsupervised cycle‐consistent deformation for shape matching
It is shown that this approach is competitive with comparable state‐of‐the‐art methods when annotated training data is readily available, but outperforms them by a large margin in the few‐shot segmentation scenario.
Deep Geometric Prior for Surface Reconstruction
This work proposes the use of a deep neural network as a geometric prior for surface reconstruction, and overfit a neural network representing a local chart parameterization to part of an input point cloud using the Wasserstein distance as a measure of approximation.
Occupancy Flow: 4D Reconstruction by Learning Particle Dynamics
This work presents Occupancy Flow, a novel spatio-temporal representation of time-varying 3D geometry with implicit correspondences which can be used for interpolation and reconstruction tasks, and believes that Occupancy flow is a promising new 4D representation which will be useful for a variety of spatio/temporal reconstruction tasks.
Learning Parallel Dense Correspondence from Spatio-Temporal Descriptors for Efficient and Robust 4D Reconstruction
This work presents a novel pipeline to learn a temporal evolution of the 3D human shape through spatially continuous transformation functions among cross-frame occupancy fields via explicitly learning continuous displacement vector fields from robust spatio-temporal shape representations.
Better Patch Stitching for Parametric Surface Reconstruction
It is shown on standard benchmarks that the use of normal consistency requirement outperforms the baselines quantitatively while enforcing better stitching leads to much better visual quality of the reconstructed objects as compared to the state-of-the-art.
Learning Local Shape Descriptors from Part Correspondences with Multiview Convolutional Networks
A new local descriptor for 3D shapes is presented, directly applicable to a wide range of shape analysis problems such as point correspondences, semantic segmentation, affordance prediction, and shape-to-scan matching by a convolutional network trained to embed geometrically and semantically similar points close to one another in descriptor space.
3D-CODED: 3D Correspondences by Deep Deformation
This work presents a new deep learning approach for matching deformable shapes by introducing Shape Deformation Networks which jointly encode 3D shapes and correspondences, and shows that this method is robust to many types of perturbations, and generalizes to non-human shapes.
Unsupervised Learning of Dense Shape Correspondence
This work introduces the first completely unsupervised correspondence learning approach for deformable 3D shapes, understanding that natural deformations approximately preserve the metric structure of the surface, yielding a natural criterion to drive the learning process toward distortion-minimizing predictions.
Shape Reconstruction by Learning Differentiable Surface Representations
This paper shows that the inherent differentiability of deep networks can be exploited to leverage differential surface properties during training so as to prevent patch collapse and strongly reduce patch overlap, and this lets us reliably compute quantities such as surface normals and curvatures.