Back to the Feature: Learning Robust Camera Localization from Pixels to Pose
This addresses the problem of camera localization for robotics and AR/VR by proposing a method that separates model parameters from scene geometry, offering incremental improvement over existing learning-based approaches.
The paper tackles camera pose estimation in known scenes by introducing PixLoc, a scene-agnostic neural network that learns robust visual features and uses geometric algorithms for pose estimation, achieving accurate 6-DoF pose from an image and 3D model with exceptional generalization to new scenes.
Camera pose estimation in known scenes is a 3D geometry task recently tackled by multiple learning algorithms. Many regress precise geometric quantities, like poses or 3D points, from an input image. This either fails to generalize to new viewpoints or ties the model parameters to a specific scene. In this paper, we go Back to the Feature: we argue that deep networks should focus on learning robust and invariant visual features, while the geometric estimation should be left to principled algorithms. We introduce PixLoc, a scene-agnostic neural network that estimates an accurate 6-DoF pose from an image and a 3D model. Our approach is based on the direct alignment of multiscale deep features, casting camera localization as metric learning. PixLoc learns strong data priors by end-to-end training from pixels to pose and exhibits exceptional generalization to new scenes by separating model parameters and scene geometry. The system can localize in large environments given coarse pose priors but also improve the accuracy of sparse feature matching by jointly refining keypoints and poses with little overhead. The code will be publicly available at https://github.com/cvg/pixloc.