arXiv subjects

Rebecca L. Russell

Publications and source records attributed to Rebecca L. Russell.

8 recordsLinked to original sources

arXiv2023

Wide-Area Geolocalization with a Limited Field of View Camera in Challenging Urban Environments

Cross-view geolocalization, a supplement or replacement for GPS, localizes an agent within a search area by matching ground-view images to overhead images. Significant progress has been made assuming a panoramic ground camera. Panoramic cameras' high complexity and cost make non-panoramic cameras more widely applicable, but also more challenging since they yield less scene overlap between ground and overhead images. This paper presents Restricted FOV Wide-Area Geolocalization (ReWAG), a cross-view geolocalization approach that combines a neural network and particle filter to globally localize a mobile agent with only odometry and a non-panoramic camera. ReWAG creates pose-aware embeddings and provides a strategy to incorporate particle pose into the Siamese network, improving localization accuracy by a factor of 100 compared to a vision transformer baseline. This extended work also presents ReWAG*, which improves upon ReWAG's generalization ability in previously unseen environments. ReWAG* repeatedly converges accurately on a dataset of images we have collected in Boston with a 72 degree field of view (FOV) camera, a location and FOV that ReWAG* was not trained on.

cs.RO↗

arXiv2023

Wide-Area Geolocalization with a Limited Field of View Camera

Cross-view geolocalization, a supplement or replacement for GPS, localizes an agent within a search area by matching images taken from a ground-view camera to overhead images taken from satellites or aircraft. Although the viewpoint disparity between ground and overhead images makes cross-view geolocalization challenging, significant progress has been made assuming that the ground agent has access to a panoramic camera. For example, our prior work (WAG) introduced changes in search area discretization, training loss, and particle filter weighting that enabled city-scale panoramic cross-view geolocalization. However, panoramic cameras are not widely used in existing robotic platforms due to their complexity and cost. Non-panoramic cross-view geolocalization is more applicable for robotics, but is also more challenging. This paper presents Restricted FOV Wide-Area Geolocalization (ReWAG), a cross-view geolocalization approach that generalizes WAG for use with standard, non-panoramic ground cameras by creating pose-aware embeddings and providing a strategy to incorporate particle pose into the Siamese network. ReWAG is a neural network and particle filter system that is able to globally localize a mobile agent in a GPS-denied environment with only odometry and a 90 degree FOV camera, achieving similar localization accuracy as what WAG achieved with a panoramic camera and improving localization accuracy by a factor of 100 compared to a baseline vision transformer (ViT) approach. A video highlight that demonstrates ReWAG's convergence on a test path of several dozen kilometers is available at https://youtu.be/U_OBQrt8qCE.

cs.RO↗

arXiv2022

Transverse charge density and the radius of the proton

A puzzling discrepancy exists between the values of the proton charge radius obtained using different experimental techniques: elastic electron-proton scattering and spectroscopy of electronic and muonic hydrogen. The proton radius is defined through the slope of the electric form factor, $G_E(Q^2)$, at zero four-momentum transfer, which is inaccessible in scattering experiments. We propose a novel method for extracting the proton radius from scattering data over a broad $Q^2$ range rather than attempting to directly determine the slope of $G_E$ at $Q^2 = 0$. This method relates the radius of the proton to its transverse charge density, which is the two-dimensional Fourier transform of the Dirac form factor, $F_1(Q^2)$. We apply our method to reanalyze the extensive data obtained by the A1 Collaboration [J. C. Bernauer et al., Phys. Rev. Lett. 105, 242001 (2010)] and extract a radius value, $r_E = 0.889(5)_{\text{stat}}(5)_{\text{syst}}(4)_{\text{model}}~\text{fm}$, that is consistent with the original result. We also provide new parametrizations for the Dirac and Pauli form factors and the transverse charge and magnetization densities of the proton. Our reanalysis shows that the proton radius discrepancy cannot be explained by issues with fitting and extrapolating the A1 data to $Q^2 = 0$.

nucl-ex↗

arXiv2021

Multivariate Uncertainty in Deep Learning

Deep learning has the potential to dramatically impact navigation and tracking state estimation problems critical to autonomous vehicles and robotics. Measurement uncertainties in state estimation systems based on Kalman and other Bayes filters are typically assumed to be a fixed covariance matrix. This assumption is risky, particularly for "black box" deep learning models, in which uncertainty can vary dramatically and unexpectedly. Accurate quantification of multivariate uncertainty will allow for the full potential of deep learning to be used more safely and reliably in these applications. We show how to model multivariate uncertainty for regression problems with neural networks, incorporating both aleatoric and epistemic sources of heteroscedastic uncertainty. We train a deep uncertainty covariance matrix model in two ways: directly using a multivariate Gaussian density loss function, and indirectly using end-to-end training through a Kalman filter. We experimentally show in a visual tracking problem the large impact that accurate multivariate uncertainty quantification can have on Kalman filter performance for both in-domain and out-of-domain evaluation data. We additionally show in a challenging visual odometry problem how end-to-end filter training can allow uncertainty predictions to compensate for filter weaknesses.

cs.LG↗

arXiv2020

A 3D Probabilistic Deep Learning System for Detection and Diagnosis of Lung Cancer Using Low-Dose CT Scans

We introduce a new computer aided detection and diagnosis system for lung cancer screening with low-dose CT scans that produces meaningful probability assessments. Our system is based entirely on 3D convolutional neural networks and achieves state-of-the-art performance for both lung nodule detection and malignancy classification tasks on the publicly available LUNA16 and Kaggle Data Science Bowl challenges. While nodule detection systems are typically designed and optimized on their own, we find that it is important to consider the coupling between detection and diagnosis components. Exploiting this coupling allows us to develop an end-to-end system that has higher and more robust performance and eliminates the need for a nodule detection false positive reduction stage. Furthermore, we characterize model uncertainty in our deep learning systems, a first for lung CT analysis, and show that we can use this to provide well-calibrated classification probabilities for both nodule detection and patient malignancy diagnosis. These calibrated probabilities informed by model uncertainty can be used for subsequent risk-based decision making towards diagnostic interventions or disease treatments, as we demonstrate using a probability-based patient referral strategy to further improve our results.

cs.CV↗

arXiv2018

Automated Vulnerability Detection in Source Code Using Deep Representation Learning

Increasing numbers of software vulnerabilities are discovered every year whether they are reported publicly or discovered internally in proprietary code. These vulnerabilities can pose serious risk of exploit and result in system compromise, information leaks, or denial of service. We leveraged the wealth of C and C++ open-source code available to develop a large-scale function-level vulnerability detection system using machine learning. To supplement existing labeled vulnerability datasets, we compiled a vast dataset of millions of open-source functions and labeled it with carefully-selected findings from three different static analyzers that indicate potential exploits. The labeled dataset is available at: https://osf.io/d45bw/. Using these datasets, we developed a fast and scalable vulnerability detection tool based on deep feature representation learning that directly interprets lexed source code. We evaluated our tool on code from both real software packages and the NIST SATE IV benchmark dataset. Our results demonstrate that deep feature representation learning on source code is a promising approach for automated software vulnerability detection.

cs.LG↗

arXiv2018

Learning to Repair Software Vulnerabilities with Generative Adversarial Networks

Motivated by the problem of automated repair of software vulnerabilities, we propose an adversarial learning approach that maps from one discrete source domain to another target domain without requiring paired labeled examples or source and target domains to be bijections. We demonstrate that the proposed adversarial learning approach is an effective technique for repairing software vulnerabilities, performing close to seq2seq approaches that require labeled pairs. The proposed Generative Adversarial Network approach is application-agnostic in that it can be applied to other problems similar to code repair, such as grammar correction or sentiment translation.

cs.CL↗

arXiv2018

Automated software vulnerability detection with machine learning

Thousands of security vulnerabilities are discovered in production software each year, either reported publicly to the Common Vulnerabilities and Exposures database or discovered internally in proprietary code. Vulnerabilities often manifest themselves in subtle ways that are not obvious to code reviewers or the developers themselves. With the wealth of open source code available for analysis, there is an opportunity to learn the patterns of bugs that can lead to security vulnerabilities directly from data. In this paper, we present a data-driven approach to vulnerability detection using machine learning, specifically applied to C and C++ programs. We first compile a large dataset of hundreds of thousands of open-source functions labeled with the outputs of a static analyzer. We then compare methods applied directly to source code with methods applied to artifacts extracted from the build process, finding that source-based models perform better. We also compare the application of deep neural network models with more traditional models such as random forests and find the best performance comes from combining features learned by deep models with tree-based models. Ultimately, our highest performing model achieves an area under the precision-recall curve of 0.49 and an area under the ROC curve of 0.87.

cs.SE↗