Daniel Cremers

TL;DR: This paper tackles the problem of data abstraction in the context of 3D point sets by using a local, low-dimensional parameterization of primitives to determine type, shape and pose of the object that a point belongs to.

TL;DR: This work proposes an object representation that combines primitive and non-primitive objects using one unified formulation that is based on signed distance fields and shows that different components of volumetric scanning, such as global trajectory optimization or geometry completion and denoising, benefit from this formulation.

TL;DR: This paper forms this matching problem as a simultaneous optimization over the entire image collection, without requiring previously computed pairwise matches to be given as input, and ensures that the final matches are consistent by construction.

TL;DR: In the theoretical contribution, it is proved that the flat metric between a parametrized current and a reference current is Lipschitz continuous in the parameters, and in experiments, the proposed shift to $k>0$ leads to interpretable and disentangled latent representations which behave equivariantly to the specified oriented tangent planes.

TL;DR: In this paper, a two-stage pipeline involves a rough yet joint estimation of orthogonal planes followed by a subsequent joint refinement of plane parameters respecting their orthogonality relations.