Visual localization is the problem of estimating the position and orientation from which a given image (or a sequence of images) is taken in a known scene. It is an important part of a wide range of computer vision and robotics applications, from self-driving cars to augmented/virtual reality systems. Visual localization techniques should work reliably and robustly under a wide range of conditions, including seasonal, weather, illumination and man-made changes. Recent benchmarking efforts model this by providing images under different conditions, and the community has made rapid progress on these datasets since their inception. However, they are limited to a few geographical regions and often recorded with a single device. We propose a new benchmark for visual localization in outdoor scenes, using crowd-sourced data to cover a wide range of geographical regions and camera devices with a focus on the failure cases of current algorithms. Experiments with state-of-the-art localization approaches show that our dataset is very challenging, with all evaluated methods failing on its hardest parts. As part of the dataset release, we provide the tooling used to generate it, enabling efficient and effective 2D correspondence annotation to obtain reference poses.

Lifelong place recognition is an essential and challenging task in computer vision with vast applications in robust localization and efficient large-scale 3D reconstruction. Progress is currently hindered by a lack of large, diverse, publicly available datasets. We contribute with Mapillary Street-Level Sequences (SLS), a large dataset for urban and suburban place recognition from image sequences. It contains more than 1.6 million images curated from the Mapillary collaborative mapping platform. The dataset is orders of magnitude larger than current data sources, and is designed to reflect the diversities of true lifelong learning. It features images from 30 major cities across six continents, hundreds of distinct cameras, and substantially different viewpoints and capture times, spanning all seasons over a nine year period. All images are geo-located with GPS and compass, and feature high-level attributes such as road type. We propose a set of benchmark tasks designed to push state-of-the-art performance and provide baseline studies. We show that current state-of-the-art methods still have a long way to go, and that the lack of diversity in existing datasets have prevented generalization to new environments. The dataset and benchmarks are available for academic research.

Single image calibration is the problem of predicting the camera parameters from one image. This problem is of importance when dealing with images collected in uncontrolled conditions by non-calibrated cameras, such as crowd-sourced applications. In this work we propose a method to predict extrinsic (tilt and roll) and intrinsic (focal length and radial distortion) parameters from a single image. We propose a parameterization for radial distortion that is better suited for learning than directly predicting the distortion parameters. Moreover, predicting additional heterogeneous variables exacerbates the problem of loss balancing. We propose a new loss function based on point projections to avoid having to balance heterogeneous loss terms. Our method is, to our knowledge, the first to jointly estimate the tilt, roll, focal length, and radial distortion parameters from a single image. We thoroughly analyze the performance of the proposed method and the impact of the improvements and compare with previous approaches for single image radial distortion correction.

In this paper, we study the problems of estimating relative pose between two cameras in the presence of radial distortion. Specifically, we consider minimal problems where one of the cameras has no or known radial distortion. There are three useful cases for this setup with a single unknown distortion: (i) fundamental matrix estimation where the two cameras are uncalibrated, (ii) essential matrix estimation for a partially calibrated camera pair, (iii) essential matrix estimation for one calibrated camera and one camera with unknown focal length. We study the parameterization of these three problems and derive fast polynomial solvers based on Gröbner basis methods. We demonstrate the numerical stability of the solvers on synthetic data. The minimal solvers have also been applied to real imagery with convincing results.

Algorithms for solving systems of polynomial equations are key components for solving geometry problems in computer vision. Fast and stable polynomial solvers are essential for numerous applications e.g. minimal problems or finding for all stationary points of certain algebraic errors. Recently, full symmetry in the polynomial systems has been utilized to simplify and speed up state-of-the-art polynomial solvers based on Gröbner basis method. In this paper, we further explore partial symmetry (i.e. where the symmetry lies in a subset of the variables) in the polynomial systems. We develop novel numerical schemes to utilize such partial symmetry. We then demonstrate the advantage of our schemes in several computer vision problems. In both synthetic and real experiments, we show that utilizing partial symmetry allow us to obtain faster and more accurate polynomial solvers than the general solvers.

In this paper, we revisit the problem of estimating the trifocal tensor from image line measurements. With measurements of corresponding lines in three views, a linear method [1] requiring 13 lines was developed to estimate the trifocal tensor from which projective reconstruction of the scene is made possible. By further utilizing the nonlinear constraints on the trifocal tensor, we propose several new linear solvers that require fewer number of lines (10,11,12) than the previous linear method. We use methods based on algebraic geometry to incorporate the non-linear constraints in the estimation. We demonstrate the performance of the proposed solvers on synthetic data. We also test the solvers on real images and obtain promising results.

Measurements with unknown time delays are common in different applications such as microphone array, radio antenna array calibration, where the sources (e.g. sounds) are transmitted in unknown time instants. In this paper, we present a method for estimating unknown time delays from Time-Difference-of-Arrival (TDOA) measurements. We propose a novel rank constraint on a matrix depending on the measurements and the unknown time delays. The time delays are recovered by solving a truncated nuclear norm minimization problem using alternating direction method of multipliers (ADMM). We show in synthetic experiments that the proposed method recovers the time delays with good accuracy for noisy and missing data.

This paper presents a complete characterization and solution to microphone position self-calibration problem for time-of-arrival (TOA) measurements. This is the problem of determining the positions of receivers and transmitters given all receiver-transmitter distances. Such calibration problems arise in application such as calibration of radio antenna networks, audio or ultra-sound arrays and WiFi transmitter arrays. We show for what cases such calibration problems are well-defined and derive efficient and numerically stable algorithms for the minimal TOA based self-calibration problems. The proposed algorithms are non-iterative and require no assumptions on the sensor positions. Experiments on synthetic data show that the minimal solvers are numerically stable and perform well on noisy data. The solvers are also tested on two real datasets with good results.

Radio based localization and tracking usually require multiple receivers/transmitters or a known floor plan. This paper presents a method for anchor free indoor positioning based on single antenna ultra wideband (UWB) measurements. By using time of arrival information from multipath propagation components stemming from scatterers with different, but unknown, positions we estimate the movement of the receiver as well as the angle of arrival of the considered multipath components. Experiments are shown for real indoor data measured in a lecture room with promising results. Simultaneous estimation of both receiver motion, transmitter and scatterer positions is performed using an factorization based approach followed by non-linear least squares optimization. A RANSAC approach to automatic matching of data has also been implemented and tested. The resulting reconstruction is compared to ground truth motion as given by the antenna positioner. The resulting accuracy is in the order of one cm.

In this paper, we study the geometry problems of estimating camera pose with unknown focal length using combination of geometric primitives. We consider points, lines and also rich features such as quivers, i.e.\ points with one or more directions. We formulate the problems as polynomial systems where the constraints for different primitives are handled in a unified way. We develop efficient polynomial solvers for each of the derived cases with different combinations of primitives. The availability of these solvers enables robust pose estimation with unknown focal length for wider classes of features. Such rich features allow for fewer feature correspondences and generate larger inlier sets with higher probability. We demonstrate in synthetic experiments that our solvers are fast and numerically stable. For real images, we show that our solvers can be used in RANSAC loops to provide good initial solutions.

In this paper, we revisit the classical perspective-n-point (PnP) problem, and propose the first non-iterative O(n) solution that is fast, generally applicable and globally optimal. Our basic idea is to formulate the PnP problem into a functional minimization problem and retrieve all its stationary points by using the Gr"obner basis technique. The novelty lies in a non-unit quaternion representation to parameterize the rotation and a simple but elegant formulation of the PnP problem into an unconstrained optimization problem. Interestingly, the polynomial system arising from its first-order optimality condition assumes two-fold symmetry, a nice property that can be utilized to improve speed and numerical stability of a Grobner basis solver. Experiment results have demonstrated that, in terms of accuracy, our proposed solution is definitely better than the state-of-the-art O(n) methods, and even comparable with the reprojection error minimization method.