Search

Completely unmanned autonomous vehicles have been anticipated for a while. Initially, these are expected to drive only under certain conditions on some roads, and advanced functionality is required to cope with the ever-increasing challenges of safety. To enhance the public's perception of road safety and trust in new vehicular technologies, we investigate in this paper the effect of several interaction paradigms with vulnerable road users by developing and applying algorithms for the automatic analysis of pedestrian body language. We assess behavioral patterns and determine the impact of the coexistence of AVs and other road users on general road safety in a shared space for VRUs and vehicles. Results showed that the implementation of visual communication cues for interacting with VRUs is not necessarily required for a shared space in which informal traffic rules apply., Comment: Published paper in the IEEE Intelligent Transportation Systems Conference - ITSC 2019

Processing of medical images such as MRI or CT presents unique challenges compared to RGB images typically used in computer vision. These include a lack of labels for large datasets, high computational costs, and metadata to describe the physical properties of voxels. Data augmentation is used to artificially increase the size of the training datasets. Training with image patches decreases the need for computational power. Spatial metadata needs to be carefully taken into account in order to ensure a correct alignment of volumes. We present TorchIO, an open-source Python library to enable efficient loading, preprocessing, augmentation and patch-based sampling of medical images for deep learning. TorchIO follows the style of PyTorch and integrates standard medical image processing libraries to efficiently process images during training of neural networks. TorchIO transforms can be composed, reproduced, traced and extended. We provide multiple generic preprocessing and augmentation operations as well as simulation of MRI-specific artifacts. Source code, comprehensive tutorials and extensive documentation for TorchIO can be found at https://torchio.rtfd.io/. The package can be installed from the Python Package Index running 'pip install torchio'. It includes a command-line interface which allows users to apply transforms to image files without using Python. Additionally, we provide a graphical interface within a TorchIO extension in 3D Slicer to visualize the effects of transforms. TorchIO was developed to help researchers standardize medical image processing pipelines and allow them to focus on the deep learning experiments. It encourages open science, as it supports reproducibility and is version controlled so that the software can be cited precisely. Due to its modularity, the library is compatible with other frameworks for deep learning with medical images., Comment: Published in Computer Methods and Programs in Biomedicine

On-board 3D object detection in autonomous vehicles often relies on geometry information captured by LiDAR devices. Albeit image features are typically preferred for detection, numerous approaches take only spatial data as input. Exploiting this information in inference usually involves the use of compact representations such as the Bird's Eye View (BEV) projection, which entails a loss of information and thus hinders the joint inference of all the parameters of the objects' 3D boxes. In this paper, we present a fully end-to-end 3D object detection framework that can infer oriented 3D boxes solely from BEV images by using a two-stage object detector and ad-hoc regression branches, eliminating the need for a post-processing stage. The method outperforms its predecessor (BirdNet) by a large margin and obtains state-of-the-art results on the KITTI 3D Object Detection Benchmark for all the categories in evaluation., Comment: Submitted to IEEE International Conference on Intelligent Transportation Systems (ITSC2020)

In this work, we prove a generalization of Quillen's Theorem A to 2-categories equipped with a special set of morphisms which we think of as weak equivalences, providing sufficient conditions for a 2-functor to induce an equivalence on $(\infty,1)$-localizations. When restricted to 1-categories with all morphisms marked, our theorem retrieves the classical Theorem A of Quillen. We additionally state and provide evidence for a new conjecture: the cofinality conjecture, which describes the relation between a conjectural theory of marked $(\infty,2)$-colimits and our generalization of Theorem A., Comment: 36 pages

The rapid development of embedded hardware in autonomous vehicles broadens their computational capabilities, thus bringing the possibility to mount more complete sensor setups able to handle driving scenarios of higher complexity. As a result, new challenges such as multiple detections of the same object have to be addressed. In this work, a siamese network is integrated into the pipeline of a well-known 3D object detector approach to suppress duplicate proposals coming from different cameras via re-identification. Additionally, associations are exploited to enhance the 3D box regression of the object by aggregating their corresponding LiDAR frustums. The experimental evaluation on the nuScenes dataset shows that the proposed method outperforms traditional NMS approaches., Comment: Submitted to IEEE Intelligent Vehicles Symposium 2020 (IV2020)

In a note published in 1925, G. H. Hardy stated the inequality \begin{equation*} \sum_{n=1}^\infty \left(\frac{1}{n}\sum_{k=1}^n a_k \right)^p \leq \left(\frac{p}{p-1}\right)^p \sum_{n=1}^\infty a_n^p, \end{equation*} for any non-negative sequence $\{a_n\}_{n \geq 1}$, and $p>1$. This inequality is known in the literature as the classical discrete Hardy inequality. It has been widely studied and several applications and new versions have been shown. In this work, we use a characterization of a weighted version of this inequality to exhibit a sufficient condition for the existence of solutions of the differential equation ${\rm div}\,{\bf u}=f$ in weighted Sobolev spaces over a certain plane irregular domain. The solvability of this equation is fundamental for the analysis of the Stokes equations. The proof follows from a local-to-global argument based on a certain decomposition of functions which is also of interest for its applications to other inequalities or related results in Sobolev spaces, such as the Korn inequality., Comment: 16 pages, 2 figures

We study the family of vertical projections whose fibers are right cosets of horizontal planes in the Heisenberg group, $\mathbb{H}^n$. We prove lower bounds for Hausdorff dimension distortion of sets under these mappings, with respect to the Euclidean metric and also the natural quotient metric which we show behaves like the Euclidean metric in this context. Our bounds are sharp in a large part of the dimension range, and we give conjectural sharp lower bounds for the remaining range. Our approach also lets us improve the known almost sure lower bound for the standard family of vertical projections in $\mathbb{H}^n$ for $n \geq 2$.

In this work, we introduce a 2-categorical variant of Lurie's relative nerve functor. We prove that it defines a right Quillen equivalence which, upon passage to $\infty$-categorical localizations, corresponds to Lurie's scaled unstraightening equivalence. In this $\infty$-bicategorical context, the relative 2-nerve provides a computationally tractable model for the Grothendieck construction which becomes equivalent, via an explicit comparison map, to Lurie's relative nerve when restricted to 1-categories., Comment: 30 pages, 1 figure, v2:minor revisions, v3: final version accepted for publication in Algebr. Geom. Topol

We obtain a Central Limit Theorem for closed Riemannian manifolds, clarifying along the way the geometric meaning of some of the hypotheses in Bhattacharya and Lin's Omnibus Central Limit Theorem for Fr\'echet means. We obtain our CLT assuming certain stability hypothesis for the cut locus, which always holds when the manifold is compact but may not be satisfied in the non-compact case., Comment: Typos corrected

This paper studies the Hausdorff dimension of the intersection of isotropic projections of subsets of $\mathbb{R}^{2n}$, as well as dimension of intersections of sets with isotropic planes. It is shown that if $A$ and $B$ are Borel subsets of $\mathbb{R}^{2n}$ of dimension greater than m, then for a positive measure set of isotropic m-planes, the intersection of the images of $A$ and $B$ under orthogonal projections onto these planes have positive Hausdorff $m$-measure. In addition, if $A$ is a measurable set of Hausdorff dimension greater than $m$, then there is a set $B\subset\mathbb{R}^{2n}$ with $\dim B\leq m$ such that for all $x\in\mathbb{R}^{2n}\setminus B$ there is a positive measure set of isotropic m-planes for which the translate by $x$ of the orthogonal complement of each such plane, intersects $A$ on a set of dimension $\dim A-m$. These results are then applied to obtain analogous results on the $n^{th}$ Heisenberg group.

Lane detection is extremely important for autonomous vehicles. For this reason, many approaches use lane boundary information to locate the vehicle inside the street, or to integrate GPS-based localization. As many other computer vision based tasks, convolutional neural networks (CNNs) represent the state-of-the-art technology to indentify lane boundaries. However, the position of the lane boundaries w.r.t. the vehicle may not suffice for a reliable positioning, as for path planning or localization information regarding lane types may also be needed. In this work, we present an end-to-end system for lane boundary identification, clustering and classification, based on two cascaded neural networks, that runs in real-time. To build the system, 14336 lane boundaries instances of the TuSimple dataset for lane detection have been labelled using 8 different classes. Our dataset and the code for inference are available online., Comment: Presented at Eurocast 2019

Many systems for autonomous vehicles' navigation rely on lane detection. Traditional algorithms usually estimate only the position of the lanes on the road, but an autonomous control system may also need to know if a lane marking can be crossed or not, and what portion of space inside the lane is free from obstacles, to make safer control decisions. On the other hand, free space detection algorithms only detect navigable areas, without information about lanes. State-of-the-art algorithms use CNNs for both tasks, with significant consumption of computing resources. We propose a novel approach that estimates the free space inside each lane, with a single CNN. Additionally, adding only a small requirement concerning GPU RAM, we infer the road type, that will be useful for path planning. To achieve this result, we train a multi-task CNN. Then, we further elaborate the output of the network, to extract polygons that can be effectively used in navigation control. Finally, we provide a computationally efficient implementation, based on ROS, that can be executed in real time. Our code and trained models are available online., Comment: Will appear in the 2019 IEEE Intelligent Vehicles Symposium (IV 2019)

Over the last years, social robots have been deployed in public environments making evident the need of human-aware navigation capabilities. In this regard, the robotics community have made efforts to include proxemics or social conventions within the navigation approaches. Nevertheless, few works have tackled the problem of labelling humans as an interactive agent when blocking the robot motion trajectory. Current state of the art navigation planners will either propose an alternative path or freeze the motion until the path is free. We present the first prototype of a framework designed to enhance social competency of robots while navigating in indoor environments. The implementation is done using Navigation and Object Detection open-source software. Specifically, the Robot Operating System (ROS) navigation stack, and OpenCV with Caffe deep learning models and MobileNet Single Shot Detector (SSD), respectively., Comment: HRI '19: ACM Workshop on Social Human-Robot Interaction of Human-care Service Robots, March 11--14, 2019, Daegu, Korea

The aim of the present study involves designing a humanoid robot guide as a walking trainer for elderly and rehabilitation patients. The system is based on the humanoid robot Pepper with a compliance approach that allows to match the motion intention of the user to the robot's pace. This feasibility study is backed up by an experimental evaluation conducted in a rehabilitation centre. We hypothesize that Pepper robot used as an assistive partner, can also benefit elderly users by motivating them to perform physical activity., Comment: 2nd Workshop on Social Robots in Therapy and Care. 14th ACM/IEEE International Conference on Human-Robot Interaction (HRI 2019)

It is currently understood that high temperature superconductivity (SC) in the transition metal $(M)$ substituted iron arsenides Ba(Fe$_{1-x}$$M$$_{x}$)$_{2}$As$_{2}$ is promoted by magnetic excitations with wave vectors $(\pi,0)$ or $(0,\pi)$. It is known that while a small amount of Co substitution leads to SC, the same does not occur for Mn for any value of $x$. In this work, magnetic excitations in the iron arsenides Ba(Fe$_{1-x}$Mn$_{x}$)$_{2}$As$_{2}$ ($x=0.0$, $0.007$, $0.009$, $0.08$) are investigated by means of Resonant Inelastic X rays Scattering (RIXS) at the Fe $L_{3}$-edge, for momentum transfer $\boldsymbol{q}$ along the high symmetry Brillouin zone $(\pi,0)$ and $(\pi,\pi)$ directions. It is shown that with increasing Mn content ($x$), the excitations become anisotropic both in dispersion and lineshape. Both effects are detected even for small values of $x$, evidencing a cooperative phenomenon between the Mn impurities, that we ascribe to emerging N\'eel order of the Mn spins. Moreover, for $x=0.08$, the excitations along $\boldsymbol{q}\parallel(\pi,0)$ are strongly damped and nearly non dispersive. This result suggests that phases of arsenides containing local moments at the FeAs layers, as in Mn or Cr substituted phases, do not support high temperature SC due to absence of the appropriate magnetic excitations., Comment: 7 pages, 5 figures, accepted in Physical Review B