Accurate localization and classification of traffic lights in driving scenes are crucial for enhancing road scene understanding in various intelligent vehicles applications. However, determining which traffic lights are relevant for the ego-vehicle remains an under-explored challenge. In this paper, we address both thelocaltask of identifying the state and relevance of each traffic light in an image and the strictly relatedglobaltask of recommending the correct course of action for the ego-vehicle (should it stop?). We propose a novel architecture, which not only localizes each traffic light and identifies its relevance with respect to the ego-vehicle, but also generates a global recommendation. To address the scarcity of datasets with these types of annotations, we introduce the Verizon Connect Traffic Light Dataset (VZC-TLD), the first U.S. dataset that provides 3,000 images annotated with traffic light boxes, states, and relevance. Experimental results on both VZC-TLD and the DriveU Traffic Light Dataset (DTLD) show that our unified approach is indeed effective, and leads to significant improvements over approaches that do not exploit the synergies between the local and global tasks. Dataset is available at:https://leotac.github.io/vzc-tld

Monocular depth estimation is a critical task for autonomous driving and many other computer vision applications. While significant progress has been made in this field, the effects of viewpoint shifts on depth estimation models remain largely underexplored. This paper introduces a novel dataset and evaluation methodology to quantify the impact of different camera positions and orientations on monocular depth estimation performance. We propose a ground truth strategy based on homography estimation and object detection, eliminating the need for expensive LIDAR sensors. We collect a diverse dataset of road scenes from multiple viewpoints and use it to assess the robustness of a modern depth estimation model to geometric shifts. After assessing the validity of our strategy on a public dataset, we provide valuable insights into the limitations of current models and highlight the importance of considering viewpoint variations in real-world applications.

Bird's Eye View (BEV) semantic maps have recently garnered a lot of attention as a useful representation of the environment to tackle assisted and autonomous driving tasks. However most of the existing work focuses on the fully supervised setting training networks on large annotated datasets. In this work we present RendBEV a new method for the self-supervised training of BEV semantic segmentation networks leveraging differentiable volumetric rendering to receive supervision from semantic perspective views computed by a 2D semantic segmentation model. Our method enables zero-shot BEV semantic segmentation and already delivers competitive results in this challenging setting. When used as pretraining to then fine-tune on labeled BEV ground truth our method significantly boosts performance in low-annotation regimes and sets a new state of the art when fine-tuning on all available labels.

The consequences of vehicle crashes are extremely costly, especially in industrial contexts, where the loss of income due to the vehicle unavailability while the incident is investigated adds to the damage produced by the event. The ongoing shift toward more connected vehicles, featuring sensors and cameras, offers the opportunity to alleviate such losses by speeding up the resolution of disputes. In this paper, we show how data routinely collected by connected vehicles can be fused to attain automatic reconstruction of the crash dynamic, a key element that has to be provided by drivers to submit a First Notification of Loss. We build upon state-of-the-art methods in areas such as SLAM, depth estimation and object detection to create a reconstruction of the scene with the vehicles involved localized both in space and time, which we present in an animated bird’s eye view. Our pipeline is evaluated on a challenging benchmark of real world videos and it is shown to create reliable reconstructions of the moment of the impact in more than 50% of scenes and overall good reconstructions in about 37% of them.

Vehicle viewpoint estimation from monocular images is a crucial component for autonomous driving vehicles and for fleet management applications. In this paper, we make several contributions to advance the state-of-the-art on this problem. We show the effectiveness of applying a smoothing filter to the output neurons of a Convolutional Neural Network (CNN) when estimating vehicle viewpoint. We point out the overlooked fact that, under the same viewpoint, the appearance of a vehicle is strongly influenced by its position in the image plane, which renders viewpoint estimation from appearance an ill-posed problem. We show how, by inserting in the model a CoordConv layer to provide the coordinates of the vehicle, we are able to solve such ambiguity and greatly increase performance. Finally, we introduce a new data augmentation technique that improves viewpoint estimation on vehicles that are closer to the camera or partially occluded. All these improvements let a lightweight CNN reach optimal results while keeping inference time low. An extensive evaluation on a viewpoint estimation benchmark (Pascal3D+) and on actual vehicle camera data (nuScenes) shows that our method significantly outperforms the state-of-the-art in vehicle viewpoint estimation, both in terms of accuracy and memory footprint.

In this article we present a novel machine learning pipeline for automatic detection of stop sign violations from dashcam videos, Inertial Measurement Units (IMU) and Global Positioning System (GPS) data. We developed a two-step approach, including a detector (Stop Sign Detector) capable of identifying stop signs presence, position, and size within video frames, followed by a classifier (Stop Violation Classifier) that assesses the presence of violations along with a severity score. The Stop Sign Detector is a deep convolutional neural network (CNN) for image classification, which leverages the information contained in its deeper layer feature maps in order to extract estimates of position and size of the detected stop signs. The Stop Violation Classifier fuses the information provided by the Stop Sign Detector with IMU/GPS data to assess the presence and severity of a stop sign violation. The proposed approach has been tested on several thousands of real-world videos, recorded from US vehicles, in all kinds of weather conditions, times of the day and environments. Our method achieves an area under the precision-recall curve of 94% with a required computational time of 2.4 seconds to process a 16-second video entirely on CPU.

The ability to detect vehicle accidents from on-board sensor data is of the utmost importance to provide prompt assistance to prevent injuries and fatalities. In this article, we present a novel deep learning method capable of analyzing time series recorded from Inertial Measurement Units (IMU) and GPS devices to recognize the presence of an accident along with its severity. We propose a neural architecture capable of exploiting the different sensor streams (i.e., acceleration, gyroscope, and GPS speed), a multimodal contrastive self-supervised training procedure, and an ad-hoc stack of data augmentation techniques, specifically designed to counteract the extreme class imbalance and to improve the generalization capabilities of the whole pipeline. The proposed method has been validated against several state-of-the-art methods on a large and highly imbalanced dataset, composed of more than 200 thousand time series collected from US vehicles, with different vehicle sizes and traveling on different types of road. Our method achieves an average-precision score (AP) of 0.9 in the detection of crashes and 0.76 in the detection of severe crashes, significantly outperforming all the other approaches, and has small footprint and latency, so that it can easily be deployed on embedded devices.

In this paper, we propose a novel deep learning architecture to classify unsafe driving maneuvers from dashcam and IMU data. Such architecture processes the output of an object detection algorithm in combination with raw video frames and GPS/IMU data. At the core of the architecture there is a novel Spatio-Temporal Attention Selector (STAS) module, which (1) extracts features describing the evolution of each object in the scene over time and (2) leverages multi-head dot product attention to select the relevant ones,i.e., the dangerous ones or the ones in danger, to perform classification. We also introduce a simple but effective methodology to increase the benefit of fine-tuning the backbone network. Our method is shown to achieve higher performance than other approaches in the literature applying attention over single frames.

BACKGROUND: Targeted amplicon sequencing of the 16S ribosomal RNA gene is one of the key tools for studying microbial diversity. The accuracy of this approach strongly depends on the choice of primer pairs and, in particular, on the balance between efficiency, specificity and sensitivity in the amplification of the different bacterial 16S sequences contained in a sample. There is thus the need for computational methods to design optimal bacterial 16S primers able to take into account the knowledge provided by the new sequencing technologies. RESULTS: We propose here a computational method for optimizing the choice of primer sets, based on multi-objective optimization, which simultaneously: 1) maximizes efficiency and specificity of target amplification; 2) maximizes the number of different bacterial 16S sequences matched by at least one primer; 3) minimizes the differences in the number of primers matching each bacterial 16S sequence. Our algorithm can be applied to any desired amplicon length without affecting computational performance. The source code of the developed algorithm is released as the mopo16S software tool (Multi-Objective Primer Optimization for 16S experiments) under the GNU General Public License and is available at http://sysbiobig.dei.unipd.it/?q=Software#mopo16S . CONCLUSIONS: Results show that our strategy is able to find better primer pairs than the ones available in the literature according to all three optimization criteria. We also experimentally validated three of the primer pairs identified by our method on multiple bacterial species, belonging to different genera and phyla. Results confirm the predicted efficiency and the ability to maximize the number of different bacterial 16S sequences matched by primers.

Motivation: A Bayesian Network is a probabilistic graphical model that encodes probabilistic dependencies between a set of random variables. We introduce bnstruct, an open source R package to (i) learn the structure and the parameters of a Bayesian Network from data in the presence of missing values and (ii) perform reasoning and inference on the learned Bayesian Networks. To the best of our knowledge, there is no other open source software that provides methods for all of these tasks, particularly the manipulation of missing data, which is a common situation in practice. Availability and Implementation: The software is implemented in R and C and is available on CRAN under a GPL licence. Contact: [email protected]. Supplementary information: Supplementary data are available at Bioinformatics online.

MOTIVATION: In the past years, both sequencing and microarray have been widely used to search for relations between genetic variations and predisposition to complex pathologies such as diabetes or neurological disorders. These studies, however, have been able to explain only a small fraction of disease heritability, possibly because complex pathologies cannot be referred to few dysfunctional genes, but are rather heterogeneous and multicausal, as a result of a combination of rare and common variants possibly impairing multiple regulatory pathways. Rare variants, though, are difficult to detect, especially when the effects of causal variants are in different directions, i.e. with protective and detrimental effects. RESULTS: Here, we propose ABACUS, an Algorithm based on a BivAriate CUmulative Statistic to identify single nucleotide polymorphisms (SNPs) significantly associated with a disease within predefined sets of SNPs such as pathways or genomic regions. ABACUS is robust to the concurrent presence of SNPs with protective and detrimental effects and of common and rare variants; moreover, it is powerful even when few SNPs in the SNP-set are associated with the phenotype. We assessed ABACUS performance on simulated and real data and compared it with three state-of-the-art methods. When ABACUS was applied to type 1 and 2 diabetes data, besides observing a wide overlap with already known associations, we found a number of biologically sound pathways, which might shed light on diabetes mechanism and etiology. AVAILABILITY AND IMPLEMENTATION: ABACUS is available at http://www.dei.unipd.it/∼dicamill/pagine/Software.html.

MOTIVATION: The increasing interest in rare genetic variants and epistatic genetic effects on complex phenotypic traits is currently pushing genome-wide association study design towards datasets of increasing size, both in the number of studied subjects and in the number of genotyped single nucleotide polymorphisms (SNPs). This, in turn, is leading to a compelling need for new methods for compression and fast retrieval of SNP data. RESULTS: We present a novel algorithm and file format for compressing and retrieving SNP data, specifically designed for large-scale association studies. Our algorithm is based on two main ideas: (i) compress linkage disequilibrium blocks in terms of differences with a reference SNP and (ii) compress reference SNPs exploiting information on their call rate and minor allele frequency. Tested on two SNP datasets and compared with several state-of-the-art software tools, our compression algorithm is shown to be competitive in terms of compression rate and to outperform all tools in terms of time to load compressed data. AVAILABILITY AND IMPLEMENTATION: Our compression and decompression algorithms are implemented in a C++ library, are released under the GNU General Public License and are freely downloadable from http://www.dei.unipd.it/~sambofra/snpack.html.

BACKGROUND: Multifactorial diseases arise from complex patterns of interaction between a set of genetic traits and the environment. To fully capture the genetic biomarkers that jointly explain the heritability component of a disease, thus, all SNPs from a genome-wide association study should be analyzed simultaneously. RESULTS: In this paper, we present Bag of Naïve Bayes (BoNB), an algorithm for genetic biomarker selection and subjects classification from the simultaneous analysis of genome-wide SNP data. BoNB is based on the Naïve Bayes classification framework, enriched by three main features: bootstrap aggregating of an ensemble of Naïve Bayes classifiers, a novel strategy for ranking and selecting the attributes used by each classifier in the ensemble and a permutation-based procedure for selecting significant biomarkers, based on their marginal utility in the classification process. BoNB is tested on the Wellcome Trust Case-Control study on Type 1 Diabetes and its performance is compared with the ones of both a standard Naïve Bayes algorithm and HyperLASSO, a penalized logistic regression algorithm from the state-of-the-art in simultaneous genome-wide data analysis. CONCLUSIONS: The significantly higher classification accuracy obtained by BoNB, together with the significance of the biomarkers identified from the Type 1 Diabetes dataset, prove the effectiveness of BoNB as an algorithm for both classification and biomarker selection from genome-wide SNP data. AVAILABILITY: Source code of the BoNB algorithm is released under the GNU General Public Licence and is available at http://www.dei.unipd.it/~sambofra/bonb.html.

The systematic perturbation of the components of a biological system has been proven among the most informative experimental setups for the identification of causal relations between the components. In this paper, we present Systematic Perturbation-Qualitative Reasoning (SPQR), a novel Qualitative Reasoning approach to automate the interpretation of the results of systematic perturbation experiments. Our method is based on a qualitative abstraction of the experimental data: for each perturbation experiment, measured values of the observed variables are modeled as lower, equal or higher than the measurements in the wild type condition, when no perturbation is applied. The algorithm exploits a set of IF-THEN rules to infer causal relations between the variables, analyzing the patterns of propagation of the perturbation signals through the biological network, and is specifically designed to minimize the rate of false positives among the inferred relations. Tested on both simulated and real perturbation data, SPQR indeed exhibits a significantly higher precision than the state of the art.

Reverse engineering is the problem of inferring the structure of a network of interactions between biological variables from a set of observations. In this paper, we propose an optimization algorithm, called MORE, for the reverse engineering of biological networks from time series data. The model inferred by MORE is a sparse system of nonlinear differential equations, complex enough to realistically describe the dynamics of a biological system. MORE tackles separately the discrete component of the problem, the determination of the biological network topology, and the continuous component of the problem, the strength of the interactions. This approach allows us both to enforce system sparsity, by globally constraining the number of edges, and to integrate a priori information about the structure of the underlying interaction network. Experimental results on simulated and real-world networks show that the mixed discrete/continuous optimization approach of MORE significantly outperforms standard continuous optimization and that MORE is competitive with the state of the art in terms of accuracy of the inferred networks.