
Learning 2D3D Correspondences To Solve The Blind PerspectivenPoint Problem
Conventional absolute camera pose via a PerspectivenPoint (PnP) solver...
read it

General techniques for approximate incidences and their application to the camera posing problem
We consider the classical camera pose estimation problem that arises in ...
read it

GloballyOptimal Inlier Set Maximisation for Simultaneous Camera Pose and Feature Correspondence
Estimating the 6DoF pose of a camera from a single image relative to a ...
read it

The Alignment of the Spheres: GloballyOptimal Spherical Mixture Alignment for Camera Pose Estimation
Determining the position and orientation of a calibrated camera from a s...
read it

Expert Sample Consensus Applied to Camera ReLocalization
Fitting model parameters to a set of noisy data points is a common probl...
read it

Learning to Find Good Correspondences
We develop a deep architecture to learn to find good correspondences for...
read it

PnPNet: A hybrid PerspectivenPoint Network
We consider the robust PerspectivenPoint (PnP) problem using a hybrid ...
read it
Solving the Blind PerspectivenPoint Problem EndToEnd With Robust Differentiable Geometric Optimization
Blind PerspectivenPoint (PnP) is the problem of estimating the position and orientation of a camera relative to a scene, given 2D image points and 3D scene points, without prior knowledge of the 2D3D correspondences. Solving for pose and correspondences simultaneously is extremely challenging since the search space is very large. Fortunately it is a coupled problem: the pose can be found easily given the correspondences and vice versa. Existing approaches assume that noisy correspondences are provided, that a good pose prior is available, or that the problem size is small. We instead propose the first fully endtoend trainable network for solving the blind PnP problem efficiently and globally, that is, without the need for pose priors. We make use of recent results in differentiating optimization problems to incorporate geometric model fitting into an endtoend learning framework, including Sinkhorn, RANSAC and PnP algorithms. Our proposed approach significantly outperforms other methods on synthetic and real data.
READ FULL TEXT
Comments
There are no comments yet.