However, a single estimate is unable to express visual ambiguity, which in many cases is unavoidable due to object symmetries or occlusion of identifying features.
The use of synthetic training data avoids this data collection problem, but a configuration of the training procedure is necessary to overcome the domain gap problem.
We present an approach to learn dense, continuous 2D-3D correspondence distributions over the surface of objects from data with no prior knowledge of visual ambiguities like symmetry.
While the use of synthetic training data prevents the need for manual annotation, there is currently a large performance gap between methods trained on real and synthetic data.
We present a learning-based method for 6 DoF pose estimation of rigid objects in point cloud data.
1 code implementation • • Tomas Hodan, Frank Michel, Eric Brachmann, Wadim Kehl, Anders Glent Buch, Dirk Kraft, Bertram Drost, Joel Vidal, Stephan Ihrke, Xenophon Zabulis, Caner Sahin, Fabian Manhardt, Federico Tombari, Tae-Kyun Kim, Jiri Matas, Carsten Rother
We propose a benchmark for 6D pose estimation of a rigid object from a single RGB-D input image.
The key insight of our work is that a single correspondence between oriented points on the two models is constrained to cast votes in a 1 DoF rotational subgroup of the full group of poses, SE(3).
We address the problem of estimating the alignment pose between two models using structure-specific local descriptors.
The novelty of our method lies in the use of a combination of local and global constraints to determine if a vote should be cast.