arXiv subjects

Andreas Wieser

Publications and source records attributed to Andreas Wieser.

32 records · Page 2Linked to original sources

arXiv2021

Simultaneous supersingular reductions of CM elliptic curves

We study the simultaneous reductions at several supersingular primes of elliptic curves with complex multiplication. We show -- under additional congruence assumptions on the CM order -- that the reductions are surjective (and even become equidistributed) on the product of supersingular loci when the discriminant of the order becomes large. This variant of the equidistribution theorems of Duke and Cornut-Vatsal is an(other) application of the recent work of Einsiedler and Lindenstrauss on the classification of joinings of higher-rank diagonalizable actions.

math.NT↗

arXiv2021

PREDATOR: Registration of 3D Point Clouds with Low Overlap

We introduce PREDATOR, a model for pairwise point-cloud registration with deep attention to the overlap region. Different from previous work, our model is specifically designed to handle (also) point-cloud pairs with low overlap. Its key novelty is an overlap-attention block for early information exchange between the latent encodings of the two point clouds. In this way the subsequent decoding of the latent representations into per-point features is conditioned on the respective other point cloud, and thus can predict which points are not only salient, but also lie in the overlap region between the two point clouds. The ability to focus on points that are relevant for matching greatly improves performance: PREDATOR raises the rate of successful registrations by more than 20% in the low-overlap scenario, and also sets a new state of the art for the 3DMatch benchmark with 89% registration recall.

cs.CV↗

arXiv2021

Planes in four space and four associated CM points

To any two-dimensional rational plane in four-dimensional space one can naturally attach a point in the Grassmannian Gr(2,4) and four lattices of rank two. Here, the first two lattices originate from the plane and its orthogonal complement and the second two essentially arise from the accidental local isomorphism between SO(4) and SO(3)xSO(3). As an application of a recent result of Einsiedler and Lindenstrauss on algebraicity of joinings we prove simultaneous equidistribution of all of these objects under two splitting conditions.

math.NT↗

arXiv2021

Weakly Supervised Learning of Rigid 3D Scene Flow

We propose a data-driven scene flow estimation algorithm exploiting the observation that many 3D scenes can be explained by a collection of agents moving as rigid bodies. At the core of our method lies a deep architecture able to reason at the \textbf{object-level} by considering 3D scene flow in conjunction with other 3D tasks. This object level abstraction, enables us to relax the requirement for dense scene flow supervision with simpler binary background segmentation mask and ego-motion annotations. Our mild supervision requirements make our method well suited for recently released massive data collections for autonomous driving, which do not contain dense scene flow annotations. As output, our model provides low-level cues like pointwise flow and higher-level cues such as holistic scene understanding at the level of rigid objects. We further propose a test-time optimization refining the predicted rigid scene flow. We showcase the effectiveness and generalization capacity of our method on four different autonomous driving datasets. We release our source code and pre-trained models under \url{github.com/zgojcic/Rigid3DSceneFlow}.

cs.CV↗

arXiv2019

The Perfect Match: 3D Point Cloud Matching with Smoothed Densities

We propose 3DSmoothNet, a full workflow to match 3D point clouds with a siamese deep learning architecture and fully convolutional layers using a voxelized smoothed density value (SDV) representation. The latter is computed per interest point and aligned to the local reference frame (LRF) to achieve rotation invariance. Our compact, learned, rotation invariant 3D point cloud descriptor achieves 94.9% average recall on the 3DMatch benchmark data set, outperforming the state-of-the-art by more than 20 percent points with only 32 output dimensions. This very low output dimension allows for near realtime correspondence search with 0.1 ms per feature point on a standard PC. Our approach is sensor- and sceneagnostic because of SDV, LRF and learning highly descriptive features with fully convolutional layers. We show that 3DSmoothNet trained only on RGB-D indoor scenes of buildings achieves 79.0% average recall on laser scans of outdoor vegetation, more than double the performance of our closest, learning-based competitors. Code, data and pre-trained models are available online at https://github.com/zgojcic/3DSmoothNet.

cs.CV↗

arXiv2019

An iterative scheme for feature based positioning using a weighted dissimilarity measure

We propose an iterative scheme for feature-based positioning using a new weighted dissimilarity measure with the goal of reducing the impact of large errors among the measured or modeled features. The weights are computed from the location-dependent standard deviations of the features and stored as part of the reference fingerprint map (RFM). Spatial filtering and kernel smoothing of the kinematically collected raw data allow efficiently estimating the standard deviations during RFM generation. In the positioning stage, the weights control the contribution of each feature to the dissimilarity measure, which in turn quantifies the difference between the set of online measured features and the fingerprints stored in the RFM. Features with little variability contribute more to the estimated position than features with high variability. Iterations are necessary because the variability depends on the location, and the location is initially unknown when estimating the position. Using real WiFi signal strength data from extended test measurements with ground truth in an office building, we show that the standard deviations of these features vary considerably within the region of interest and are neither simple functions of the signal strength nor of the distances from the corresponding access points. This is the motivation to include the empirical standard deviations in the RFM. We then analyze the deviations of the estimated positions with and without the location-dependent weighting. In the present example the maximum radial positioning error from ground truth are reduced by 40% comparing to kNN without the weighted dissimilarity measure.

cs.LG↗

arXiv2019

Linnik's problems and maximal entropy methods

We use maximal entropy methods to examine the distribution properties of primitive integer points on spheres and of CM points on the modular surface. The proofs we give are a modern and dynamical interpretation of Linnik's original ideas and follow techniques presented by Einsiedler, Lindenstrauss, Michel and Venkatesh in 2011.

math.DS↗

arXiv2019

Modified Jaccard Index Analysis and Adaptive Feature Selection for Location Fingerprinting with Limited Computational Complexity

We propose an approach for fingerprinting-based positioning which reduces the data requirements and computational complexity of the online positioning stage. It is based on a segmentation of the entire region of interest into subregions, identification of candidate subregions during the online-stage, and position estimation using a preselected subset of relevant features. The subregion selection uses a modified Jaccard index which quantifies the similarity between the features observed by the user and those available within the reference fingerprint map. The adaptive feature selection is achieved using an adaptive forward-backward greedy search which determines a subset of features for each subregion, relevant with respect to a given fingerprinting-based positioning method. In an empirical study using signals of opportunity for fingerprinting the proposed subregion and feature selection reduce the processing time during the online-stage by a factor of about 10 while the positioning accuracy does not deteriorate significantly. In fact, in one of the two study cases the 90th percentile of the circular error increased by 7.5% while in the other study case we even found a reduction of the corresponding circular error by 30%.

stat.AP↗

arXiv2018

CDM: Compound dissimilarity measure and an application to fingerprinting-based positioning

A non-vector-based dissimilarity measure is proposed by combining vector-based distance metrics and set operations. This proposed compound dissimilarity measure (CDM) is applicable to quantify similarity of collections of attribute/feature pairs where not all attributes are present in all collections. This is a typical challenge in the context of e.g., fingerprinting-based positioning (FbP). Compared to vector-based distance metrics (e.g., Minkowski), the merits of the proposed CDM are i) the data do not need to be converted to vectors of equal dimension, ii) shared and unshared attributes can be weighted differently within the assessment, and iii) additional degrees of freedom within the measure allow to adapt its properties to application needs in a data-driven way. We indicate the validity of the proposed CDM by demonstrating the improvements of the positioning performance of fingerprinting-based WLAN indoor positioning using four different datasets, three of them publicly available. When processing these datasets using CDM instead of conventional distance metrics the accuracy of identifying buildings and floors improves by about 5% on average. The 2d positioning errors in terms of root mean squared error (RMSE) are reduced by a factor of two, and the percentage of position solutions with less than 2m error improves by over 10%.

stat.AP↗

arXiv2017

Jaccard analysis and LASSO-based feature selection for location fingerprinting with limited computational complexity

We propose an approach to reduce both computational complexity and data storage requirements for the online positioning stage of a fingerprinting-based indoor positioning system (FIPS) by introducing segmentation of the region of interest (RoI) into sub-regions, sub-region selection using a modified Jaccard index, and feature selection based on randomized least absolute shrinkage and selection operator (LASSO). We implement these steps into a Bayesian framework of position estimation using the maximum a posteriori (MAP) principle. An additional benefit of these steps is that the time for estimating the position, and the required data storage are virtually independent of the size of the RoI and of the total number of available features within the RoI. Thus the proposed steps facilitate application of FIPS to large areas. Results of an experimental analysis using real data collected in an office building using a Nexus 6P smart phone as user device and a total station for providing position ground truth corroborate the expected performance of the proposed approach. The positioning accuracy obtained by only processing 10 automatically identified features instead of all available ones and limiting position estimation to 10 automatically identified sub-regions instead of the entire RoI is equivalent to processing all available data. In the chosen example, 50% of the errors are less than 1.8 m and 90% are less than 5 m. However, the computation time using the automatically identified subset of data is only about 1% of that required for processing the entire data set.

stat.ML↗

arXiv2017

WiFi based trajectory alignment, calibration and easy site survey using smart phones and foot-mounted IMUs

Foot-mounted inertial positioning (FMIP) can face problems of inertial drifts and unknown initial states in real applications, which renders the estimated trajectories inaccurate and not obtained in a well defined coordinate system for matching trajectories of different users. In this paper, an approach adopting received signal strength (RSS) measurements for Wifi access points (APs) are proposed to align and calibrate the trajectories estimated from foot mounted inertial measurement units (IMUs). A crowd-sourced radio map (RM) can be built subsequently and can be used for fingerprinting based Wifi indoor positioning (FWIP). The foundation of the proposed approach is graph based simultaneously localization and mapping (SLAM). The nodes in the graph denote users poses and the edges denote the pairwise constrains between the nodes. The constrains are derived from: (1) inertial estimated trajectories; (2) vicinity in the RSS space. With these constrains, an error functions is defined. By minimizing the error function, the graph is optimized and the aligned/calibrated trajectories along with the RM are acquired. The experimental results have corroborated the effectiveness of the approach for trajectory alignment, calibration as well as RM construction.

stat.ML↗

arXiv2017

Pedestrian Positioning Using WiFi Fingerprints and a Foot-mounted Inertial Sensor

Foot-mounted inertial positioning (FMIP) and fingerprinting based WiFi indoor positioning (FWIP) are two promising solutions for indoor positioning. However, FMIP suffers from accumulative positioning errors in the long term while FWIP involves a very labor-intensive offline training phase. A new approach combining the two solutions is proposed in this paper, which can limit the error growth in FMIP and is free of any offline site survey phase. This approach is realized in the framework of a particle filter, where each particle denotes a potential trajectory of the user and is weighted according to its consistency in signal strength space. Compared with the traditional Gaussian process based approaches, the proposed one has less computational cost and is free from any prior information in the position domain, such as the positions of access points, received signal strengths at certain positions and so on. An experiment is carried out to demonstrate the performance of the proposed approach compared to the traditional Gaussian process based approach.

stat.AP↗

arXiv2017

Fast Radio Map Construction and Position Estimation via Direct Mapping for WLAN Indoor Localization System

The main limitation that constrains the fast and comprehensive application of Wireless Local Area Network (WLAN) based indoor localization systems with Received Signal Strength (RSS) positioning algorithms is the building of the fingerprinting radio map, which is time-consuming especially when the indoor environment is large and/or with high frequent changes. Different approaches have been proposed to reduce workload, including fingerprinting deployment and update efforts, but the performance degrades greatly when the workload is reduced below a certain level. In this paper, we propose an indoor localization scenario that applies metric learning and manifold alignment to realize direct mapping localization (DML) using a low resolution radio map with single sample of RSS that reduces the fingerprinting workload by up to 87\%. Compared to previous work. The proposed two localization approaches, DML and $k$ nearest neighbors based on reconstructed radio map (reKNN), were shown to achieve less than 4.3\ m and 3.7\ m mean localization error respectively in a typical office environment with an area of approximately 170\ m$^2$, while the unsupervised localization with perturbation algorithm was shown to achieve 4.7\ m mean localization error with 8 times more workload than the proposed methods. As for the room level localization application, both DML and reKNN can meet the requirement with at most 9\ m of localization error which is enough to tell apart different rooms with over 99\% accuracy.

cs.NI↗

arXiv2017

Application of backpropagation neural networks to both stages of fingerprinting based WIPS

We propose a scheme to employ backpropagation neural networks (BPNNs) for both stages of fingerprinting-based indoor positioning using WLAN/WiFi signal strengths (FWIPS): radio map construction during the offline stage, and localization during the online stage. Given a training radio map (TRM), i.e., a set of coordinate vectors and associated WLAN/WiFi signal strengths of the available access points, a BPNN can be trained to output the expected signal strengths for any input position within the region of interest (BPNN-RM). This can be used to provide a continuous representation of the radio map and to filter, densify or decimate a discrete radio map. Correspondingly, the TRM can also be used to train another BPNN to output the expected position within the region of interest for any input vector of recorded signal strengths and thus carry out localization (BPNN-LA).Key aspects of the design of such artificial neural networks for a specific application are the selection of design parameters like the number of hidden layers and nodes within the network, and the training procedure. Summarizing extensive numerical simulations, based on real measurements in a testbed, we analyze the impact of these design choices on the performance of the BPNN and compare the results in particular to those obtained using the $k$ nearest neighbors ($k$NN) and weighted $k$ nearest neighbors approaches to FWIPS.

stat.ML↗