Search

Semantic segmentation on LiDAR imaging is increasingly gaining attention, as it can provide useful knowledge for perception systems and potential for autonomous driving. However, collecting and labeling real LiDAR data is an expensive and time-consuming task. While datasets such as SemanticKITTI have been manually collected and labeled, the introduction of simulation tools such as CARLA, has enabled the creation of synthetic datasets on demand. In this work, we present a modified CARLA simulator designed with LiDAR semantic segmentation in mind, with new classes, more consistent object labeling with their counterparts from real datasets such as SemanticKITTI, and the possibility to adjust the object class distribution. Using this tool, we have generated SynthmanticLiDAR, a synthetic dataset for semantic segmentation on LiDAR imaging, designed to be similar to SemanticKITTI, and we evaluate its contribution to the training process of different semantic segmentation algorithms by using a naive transfer learning approach. Our results show that incorporating SynthmanticLiDAR into the training process improves the overall performance of tested algorithms, proving the usefulness of our dataset, and therefore, our adapted CARLA simulator. The dataset and simulator are available in https://github.com/vpulab/SynthmanticLiDAR., Comment: 2024 IEEE International Conference on Image Processing (ICIP)

Pollution that originates from automobiles is a concern in the current world, not only because of global warming, but also due to the harmful effects on people's health and lives. Despite regulations on exhaust gas emissions being applied, minimizing unsuitable driving habits that cause elevated fuel consumption and emissions would achieve further reductions. For that reason, this work proposes a self-organized map (SOM)-based intelligent system in order to provide drivers with eco-driving-intended driving style (DS) recommendations. The development of the DS advisor uses driving data from the Uyanik instrumented car. The system classifies drivers regarding the underlying causes of non-optimal DSs from the eco-driving viewpoint. When compared with other solutions, the main advantage of this approach is the personalization of the recommendations that are provided to motorists, comprising the handling of the pedals and the gearbox, with potential improvements in both fuel consumption and emissions ranging from the 9.5\% to the 31.5\%, or even higher for drivers that are strongly engaged with the system. It was successfully implemented using a field-programmable gate array (FPGA) device of the Xilinx ZynQ programmable system-on-a-chip (PSoC) family. This SOM-based system allows for real-time implementation, state-of-the-art timing performances, and low power consumption, which are suitable for developing advanced driving assistance systems (ADASs).

Advanced driving assistance systems (ADAS) are primarily designed to increase driving safety and reduce traffic congestion without paying too much attention to passenger comfort or motion sickness. However, in view of autonomous cars, and taking into account that the lack of comfort and motion sickness increase in passengers, analysis from a comfort perspective is essential in the future car investigation. The aim of this work is to study in detail how passenger's comfort evaluation parameters vary depending on the driving style, car or road. The database used has been developed by compiling the accelerations suffered by passengers when three drivers cruise two different vehicles on different types of routes. In order to evaluate both comfort and motion sickness, first, the numerical values of the main comfort evaluation variables reported in the literature have been analyzed. Moreover, a complementary statistical analysis of probability density and a power spectral analysis are performed. Finally, quantitative results are compared with passenger qualitative feedback. The results show the high dependence of comfort evaluation variables' value with the road type. In addition, it has been demonstrated that the driving style and vehicle dynamics amplify or attenuate those values. Additionally, it has been demonstrated that contributions from longitudinal and lateral accelerations have a much greater effect in the lack of comfort than vertical ones. Finally, based on the concrete results obtained, a new experimental campaign is proposed.

New challenges on transport systems are emerging due to the advances that the current paradigm is experiencing. The breakthrough of the autonomous car brings concerns about ride comfort, while the pollution concerns have arisen in recent years. In the model of automated automobiles, drivers are expected to become passengers, so, they will be more prone to suffer from ride discomfort or motion sickness. Conversely, the eco-driving implications should not be set aside because of the influence of pollution on climate and people's health. For that reason, a joint assessment of the aforementioned points would have a positive impact. Thus, this work presents a self-organised map-based solution to assess ride comfort features of individuals considering their driving style from the viewpoint of eco-driving. For this purpose, a previously acquired dataset from an instrumented car was used to classify drivers regarding the causes of their lack of ride comfort and eco-friendliness. Once drivers are classified regarding their driving style, natural-language-based recommendations are proposed to increase the engagement with the system. Hence, potential improvements of up to the 57.7% for ride comfort evaluation parameters, as well as up to the 47.1% in greenhouse-gasses emissions are expected to be reached.

Advanced driving-assistance systems (ADAS) are intended to automatize driver tasks, as well as improve driving and vehicle safety. This work proposes an intelligent neuro-fuzzy sensor for driving style (DS) recognition, suitable for ADAS enhancement. The development of the driving style intelligent sensor uses naturalistic driving data from the SHRP2 study, which includes data from a CAN bus, inertial measurement unit, and front radar. The system has been successfully implemented using a field-programmable gate array (FPGA) device of the Xilinx Zynq programmable system-on-chip (PSoC). It can mimic the typical timing parameters of a group of drivers as well as tune these typical parameters to model individual DSs. The neuro-fuzzy intelligent sensor provides high-speed real-time active ADAS implementation and is able to personalize its behavior into safe margins without driver intervention. In particular, the personalization procedure of the time headway (THW) parameter for an ACC in steady car following was developed, achieving a performance of 0.53 microseconds. This performance fulfilled the requirements of cutting-edge active ADAS specifications., Comment: Journal Article

Semantic segmentation is a computer vision task where classification is performed at a pixel level. Due to this, the process of labeling images for semantic segmentation is time-consuming and expensive. To mitigate this cost there has been a surge in the use of synthetically generated data -- usually created using simulators or videogames -- which, in combination with domain adaptation methods, can effectively learn how to segment real data. Still, these datasets have a particular limitation: due to their closed-set nature, it is not possible to include novel classes without modifying the tool used to generate them, which is often not public. Concurrently, generative models have made remarkable progress, particularly with the introduction of diffusion models, enabling the creation of high-quality images from text prompts without additional supervision. In this work, we propose an unsupervised pipeline that leverages Stable Diffusion and Segment Anything Module to generate class examples with an associated segmentation mask, and a method to integrate generated cutouts for novel classes in semantic segmentation datasets, all with minimal user input. Our approach aims to improve the performance of unsupervised domain adaptation methods by introducing novel samples into the training data without modifications to the underlying algorithms. With our methods, we show how models can not only effectively learn how to segment novel classes, with an average performance of 51% IoU, but also reduce errors for other, already existing classes, reaching a higher performance level overall.

This paper introduces a novel synthetic dataset that captures urban scenes under a variety of weather conditions, providing pixel-perfect, ground-truth-aligned images to facilitate effective feature alignment across domains. Additionally, we propose a method for domain adaptation and generalization that takes advantage of the multiple versions of each scene, enforcing feature consistency across different weather scenarios. Our experimental results demonstrate the impact of our dataset in improving performance across several alignment metrics, addressing key challenges in domain adaptation and generalization for segmentation tasks. This research also explores critical aspects of synthetic data generation, such as optimizing the balance between the volume and variability of generated images to enhance segmentation performance. Ultimately, this work sets forth a new paradigm for synthetic data generation and domain adaptation.

Foundation models trained on web-scraped datasets propagate societal biases to downstream tasks. While counterfactual generation enables bias analysis, existing methods introduce artifacts by modifying contextual elements like clothing and background. We present a localized counterfactual generation method that preserves image context by constraining counterfactual modifications to specific attribute-relevant regions through automated masking and guided inpainting. When applied to the Conceptual Captions dataset for creating gender counterfactuals, our method results in higher visual and semantic fidelity than state-of-the-art alternatives, while maintaining the performance of models trained using only real data on non-human-centric tasks. Models fine-tuned with our counterfactuals demonstrate measurable bias reduction across multiple metrics, including a decrease in gender classification disparity and balanced person preference scores, while preserving ImageNet zero-shot performance. The results establish a framework for creating balanced datasets that enable both accurate bias profiling and effective mitigation.

Most of current computer vision-based advanced driver assistance systems (ADAS) perform detection and tracking of objects quite successfully under regular conditions. However, under adverse weather and changing lighting conditions, and in complex situations with many overlapping objects, these systems are not completely reliable. The spectral reflectance of the different objects in a driving scene beyond the visible spectrum can offer additional information to increase the reliability of these systems, especially under challenging driving conditions. Furthermore, this information may be significant enough to develop vision systems that allow for a better understanding and interpretation of the whole driving scene. In this work we explore the use of snapshot, video-rate hyperspectral imaging (HSI) cameras in ADAS on the assumption that the near infrared (NIR) spectral reflectance of different materials can help to better segment the objects in real driving scenarios. To do this, we have used the HSI-Drive 1.1 dataset to perform various experiments on spectral classification algorithms. However, the information retrieval of hyperspectral recordings in natural outdoor scenarios is challenging, mainly because of deficient colour constancy and other inherent shortcomings of current snapshot HSI technology, which poses some limitations to the development of pure spectral classifiers. In consequence, in this work we analyze to what extent the spatial features codified by standard, tiny fully convolutional network (FCN) models can improve the performance of HSI segmentation systems for ADAS applications. The abstract above is truncated due to submission limits. For the full abstract, please refer to the published article.

The article discusses the use of low cost System-On-Module (SOM) platforms for the implementation of efficient hyperspectral imaging (HSI) processors for application in autonomous driving. The work addresses the challenges of shaping and deploying multiple layer fully convolutional networks (FCN) for low-latency, on-board image semantic segmentation using resource- and power-constrained processing devices. The paper describes in detail the steps followed to redesign and customize a successfully trained HSI segmentation lightweight FCN that was previously tested on a high-end heterogeneous multiprocessing system-on-chip (MPSoC) to accommodate it to the constraints imposed by a low-cost SOM. This SOM features a lower-end but much cheaper MPSoC suitable for the deployment of automatic driving systems (ADS). In particular the article reports the data- and hardware-specific quantization techniques utilized to fit the FCN into a commercial fixed-point programmable AI coprocessor IP, and proposes a full customized post-training quantization scheme to reduce computation and storage costs without compromising segmentation accuracy.

The scarcity of data acquired under actual space operational conditions poses a significant challenge for developing learning-based visual navigation algorithms crucial for autonomous spacecraft navigation. This data shortage is primarily due to the prohibitive costs and inherent complexities of space operations. While existing datasets, predominantly relying on computer-simulated data, have partially addressed this gap, they present notable limitations. Firstly, these datasets often utilize proprietary image generation tools, restricting the evaluation of navigation methods in novel, unseen scenarios. Secondly, they provide limited ground-truth data, typically focusing solely on the spacecraft's translation and rotation relative to the camera. To address these limitations, we present SPIN (SPacecraft Imagery for Navigation), an open-source spacecraft image generation tool designed to support a wide range of visual navigation scenarios in space, with a particular focus on relative navigation tasks. SPIN provides multiple modalities of ground-truth data and allows researchers to employ custom 3D models of satellites, define specific camera-relative poses, and adjust settings such as camera parameters or environmental illumination conditions. We also propose a method for exploiting our tool as a data augmentation module. We validate our tool on the spacecraft pose estimation task by training with a SPIN-generated replica of SPEED+, reaching a 47% average error reduction on SPEED+ testbed data (that simulates realistic space conditions), further reducing it to a 60% error reduction when using SPIN as a data augmentation method. Both the SPIN tool (and source code) and our SPIN-generated version of SPEED+ will be publicly released upon paper acceptance on GitHub. https://github.com/vpulab/SPIN

Deep learning technologies have already demonstrated a high potential to build diagnosis support systems from medical imaging data, such as Chest X-Ray images. However, the shortage of labeled data in the medical field represents one key obstacle to narrow down the performance gap with respect to applications in other image domains. In this work, we investigate the benefits of a curricular Self-Supervised Learning (SSL) pretraining scheme with respect to fully-supervised training regimes for pneumonia recognition on Chest X-Ray images of Covid-19 patients. We show that curricular SSL pretraining, which leverages unlabeled data, outperforms models trained from scratch, or pretrained on ImageNet, indicating the potential of performance gains by SSL pretraining on massive unlabeled datasets. Finally, we demonstrate that top-performing SSLpretrained models show a higher degree of attention in the lung regions, embodying models that may be more robust to possible external confounding factors in the training datasets, identified by previous works.

The confrontation between the superconducting fluctuations and percolation effects as the origin of the in-plane paraconductivity in cuprate superconductors was earlier addressed at a quantitative level in the case of the optimally doped YBa$_2$Cu$_3$O$_{7-\delta}$ (YBCO) compound. Using in-plane resistivity data from a high-quality YBCO thin film, we will extend these analyses to high reduced temperatures, in the case of the Gaussian-Ginzburg-Landau (GGL) approach for the conventional superconducting fluctuations, by considering the total energy cutoff. These data will also be analysed in terms of the mean field-approach of the effective-medium theory, to probe if emergent percolative effects may account for the resistivity rounding above $T_c$. Our analyses confirm earlier conclusions: the measured paraconductivity cannot be explained in terms of emergent percolation processes, but it may be accounted for in terms of the GGL approach. These results also call into question alternative scenarios, including a recent proposal derived from emergent percolative effects., Comment: 14 pages including 3 figures

Deep neural networks are efficient at learning the data distribution if it is sufficiently sampled. However, they can be strongly biased by non-relevant factors implicitly incorporated in the training data. These include operational biases, such as ineffective or uneven data sampling, but also ethical concerns, as the social biases are implicitly present\textemdash even inadvertently, in the training data or explicitly defined in unfair training schedules. In tasks having impact on human processes, the learning of social biases may produce discriminatory, unethical and untrustworthy consequences. It is often assumed that social biases stem from supervised learning on labelled data, and thus, Self-Supervised Learning (SSL) wrongly appears as an efficient and bias-free solution, as it does not require labelled data. However, it was recently proven that a popular SSL method also incorporates biases. In this paper, we study the biases of a varied set of SSL visual models, trained using ImageNet data, using a method and dataset designed by psychological experts to measure social biases. We show that there is a correlation between the type of the SSL model and the number of biases that it incorporates. Furthermore, the results also suggest that this number does not strictly depend on the model's accuracy and changes throughout the network. Finally, we conclude that a careful SSL model selection process can reduce the number of social biases in the deployed model, whilst keeping high performance., Comment: 10 pages, 6 figures, accepted in CVPR2022

Cross-camera image data association is essential for many multi-camera computer vision tasks, such as multi-camera pedestrian detection, multi-camera multi-target tracking, 3D pose estimation, etc. This association task is typically stated as a bipartite graph matching problem and often solved by applying minimum-cost flow techniques, which may be computationally inefficient with large data. Furthermore, cameras are usually treated by pairs, obtaining local solutions, rather than finding a global solution at once. Other key issue is that of the affinity measurement: the widespread usage of non-learnable pre-defined distances, such as the Euclidean and Cosine ones. This paper proposes an efficient approach for cross-cameras data-association focused on a global solution, instead of processing cameras by pairs. To avoid the usage of fixed distances, we leverage the connectivity of Graph Neural Networks, previously unused in this scope, using a Message Passing Network to jointly learn features and similarity. We validate the proposal for pedestrian multi-view association, showing results over the EPFL multi-camera pedestrian dataset. Our approach considerably outperforms the literature data association techniques, without requiring to be trained in the same scenario in which it is tested. Our code is available at \url{http://www-vpu.eps.uam.es/publications/gnn_cca}.

State-of-the-art deep learning approaches for skin lesion recognition often require pretraining on larger and more varied datasets, to overcome the generalization limitations derived from the reduced size of the skin lesion imaging datasets. ImageNet is often used as the pretraining dataset, but its transferring potential is hindered by the domain gap between the source dataset and the target dermatoscopic scenario. In this work, we introduce a novel pretraining approach that sequentially trains a series of Self-Supervised Learning pretext tasks and only requires the unlabeled skin lesion imaging data. We present a simple methodology to establish an ordering that defines a pretext task curriculum. For the multi-class skin lesion classification problem, and ISIC-2019 dataset, we provide experimental evidence showing that: i) a model pretrained by a curriculum of pretext tasks outperforms models pretrained by individual pretext tasks, and ii) a model pretrained by the optimal pretext task curriculum outperforms a model pretrained on ImageNet. We demonstrate that this performance gain is related to the fact that the curriculum of pretext tasks better focuses the attention of the final model on the skin lesion. Beyond performance improvement, this strategy allows for a large reduction in the training time with respect to ImageNet pretraining, which is especially advantageous for network architectures tailored for a specific problem., Comment: 11 pages, 8 figures, submitted to the Journal of Biomedical and Health Informatics (Special Issue on Skin Image Analysis in the Age of Deep Learning)

Detailed measurements of the in-plane resistivity were performed in a high-quality Ba(Fe$_{1-x}$Co$_{x}$)$_2$As$_2$ ($x=0.065$) single crystal, in magnetic fields up to 9 T and with different orientations $\theta$ relative to the crystal $c$ axis. A significant $\rho(T)_{H,\theta}$ rounding is observed just above the superconducting critical temperature $T_c$ due to Cooper pairs created by superconducting fluctuations. These data are analyzed in terms of a generalization of the Aslamazov-Larkin approach, that extends its applicability to high reduced-temperatures and magnetic fields. This method allows us to carry out a criterion-independent determination of the angular dependence of the upper critical field, $H_{c2}(\theta)$. In spite of the relatively small anisotropy of this compound, it is found that $H_{c2}(\theta)$ presents a significant deviation from the single-band 3D anisotropic Ginzburg-Landau (3D-aGL) approach, particularly for large $\theta$ (typically above $\sim60^o$). These results are interpreted in terms of the multiband nature of these materials, in contrast with other proposals for similar $H_{c2}(\theta)$ anomalies. Our results are also consistent with an effective anisotropy factor almost temperature independent near $T_c$, a result that differs from the ones obtained by using a single-band model., Comment: 14 pages including 7 figures. This is a preprint of an article published in Scientific Reports. The final authenticated version is available online at: https://rdcu.be/clIHJ