Semantic scene completion (SSC) aims to infer both the 3D geometry and semantics of a scene from single images. In contrast to prior work on SSC that heavily relies on expensive ground-truth annotations, we approach SSC in an unsupervised setting. Our novel method, SceneDINO, adapts techniques from self-supervised representation learning and 2D unsupervised scene understanding to SSC. Our training exclusively utilizes multi-view consistency self-supervision without any form of semantic or geometric ground truth. Given a single input image, SceneDINO infers the 3D geometry and expressive 3D DINO features in a feed-forward manner. Through a novel 3D feature distillation approach, we obtain unsupervised 3D semantics. In both 3D and 2D unsupervised scene understanding, SceneDINO reaches state-of-the-art segmentation accuracy. Linear probing our 3D features matches the segmentation accuracy of a current supervised SSC approach. Additionally, we showcase the domain generalization and multi-view consistency of SceneDINO, taking the first steps towards a strong foundation for single image 3D scene understanding.

The creation of digital twins of cityscapes requires the understanding the semantics of relevant objects encountered in the scene, with classes possibly not well covered in standard datasets. In this paper, we assess the generalization capabilities of models for semantic segmentation of street scenes trained on synthetic data to real images. In order to perform assessment on real data, we create a dataset from images obtained from a vehicle-based capture setup, and propose a highly automated annotation pipeline for generating ground truth. We train three state of the art segmentation models on our own synthetic dataset, and evaluate the methods on real data.

Introduction: Sepsis is a life-threatening condition characterized by organ dysfunction evoked by an abnormal host response to an infectious process. Statins, a class of lipid-lowering agents, possess immunomodulatory effects. Having in mind the potential benefit of treatment with statins in sepsis, we aimed to investigate the effects of statins on cardiac function and cardiac oxidative stress (OS) in the experimental sepsis rat model. Methodology: Thirty Wistar Albino rats (males, 8-weeks old, BW were randomly divided into 2 groups: Control group (C,) and animals with induced sepsis (SEPSIS,). After induced sepsis, all animals from the SEPSIS group were randomly divided into 4 subgroups: animals with induced sepsis without treatment (S), animals with induced sepsis treated with a single dose of Atorvastatin, Simvastatin, and Rosuvastatin. After 72h from sepsis induction, all animals were sacrificed and hearts were isolated and perfused according to the Langendorff technique gradually increasing coronary perfusion pressures 40–120 cmH20. In coronary venous effluent were determined the biomarkers of cardiac OS. All data were analyzed by one-way Anova and Kruskal-Wallis tests$(p < 0.05)$. Results: There were no changes in cardiodynamic parameters in septic rats without treatment, as well in those treated with statins, except for coronary flow, where its values were statistically increased in the sepsis group compared to all other groups. Also, sepsis was associated with disturbed cardiac OS and most of the applied therapeutic protocols of statins have mitigated the release of hydrogen peroxide and lipid peroxidation index, while in the case of superoxide anion radical, only atorvastatin at CPP 120 cmH2O had such a positive effect. Conclusion: Given the conditions of the current experiment, we can generally conclude that the rat's hearts in septic conditions were exposed to elevated OS which was not related to its functional changes. Additionally, statin therapy has achieved positive effects in terms of reduced release of molecules that could cause oxidative damage to the rat heart.

Interaction of socially assistive robots with users is based on social cues coming from different interaction modalities, such as speech or gestures. However, using all modalities at all times may be inefficient as it can overload the user with redundant information and increase the task completion time. Additionally, users may favor certain modalities over the other as a result of their disability or personal preference. In this paper, we propose an Adaptive Modality Selection (AMS) algorithm that chooses modalities depending on the state of the user and the environment, as well as user preferences. The variables that describe the environment and the user state are defined as resources, and we posit that modalities are successful if certain resources possess specific values during their use. Besides the resources, the proposed algorithm takes into account user preferences which it learns while interacting with users. We tested our algorithm in simulations, and we implemented it on a robotic system that provides cognitive training, specifically Sequential memory exercises. Experimental results show that it is possible to use only a subset of available modalities without compromising the interaction. Moreover, we see a trend for users to perform better when interacting with a system with implemented AMS algorithm.

In many complex robotics systems, interaction takes place in all directions between human, robot, and environment. Performance of such a system depends on this interaction, and a proper evaluation of a system must build on a proper modeling of interaction, a relevant set of performance metrics, and a methodology to combine metrics into a single performance value. In this paper, existing models of human-robot interaction are adapted to fit complex scenarios with one or several humans and robots. The interaction and the evaluation process is formalized, and a general method to fuse performance values over time and for several performance metrics is presented. The resulting value, denoted interaction quality, adds a dimension to ordinary performance metrics by being explicit about the interplay between performance metrics, and thereby provides a formal framework to understand, model, and address complex aspects of evaluation of human-robot interaction.

Robots that can assist in activities of daily living (ADL) such as dressing assistance, need to be capable of intuitive and safe interaction. Vision systems are often used to provide information on the position and movement of the robot and user. However, in a dressing context, technical complexity, occlusion and concerns over user privacy pushes research to investigate other approaches for human-robot interaction (HRI). We analysed verbal, proprioceptive and force feedback from 18 participants during a human-human dressing experiment where users received dressing assistance from a researcher mimicking robot behaviour. This paper investigates the occurrence of deictic speech in an assisted-dressing task and how any ambiguity could be resolved to ensure safe and reliable HRI. We focus on one of the most frequently occurring deictic words “up”, which was captured over 300 times during the experiments and is used as an example of an ambiguous command. We attempt to resolve the ambiguity of these commands through predictive models. These models were used to predict end effector choice and the direction in which the garment should move. The model for predicting end effector choice resulted in 70.4% accuracy based on the user's head orientation. For predicting garment direction, the model used the angle of the user's arm and resulted in 87.8% accuracy. We also found that additional categories such as the starting position of the user's arms and end-effector height may improve the accuracy of a predictive model. We present suggestions on how these inputs may be attained through non-visual means, for example through haptic perception of end-effector position, proximity sensors and acoustic source localisation.

Three advanced natural interaction modalities for mobile robot guidance in an indoor environment were developed and compared using two tasks and quantitative metrics to measure performance and workload. The first interaction modality is based on direct physical interaction requiring the human user to push the robot in order to displace it. The second and third interaction modalities exploit a 3-D vision-based human-skeleton tracking allowing the user to guide the robot by either walking in front of it or by pointing toward a desired location. In the first task, the participants were asked to guide the robot between different rooms in a simulated physical apartment requiring rough movement of the robot through designated areas. The second task evaluated robot guidance in the same environment through a set of waypoints, which required accurate movements. The three interaction modalities were implemented on a generic differential drive mobile platform equipped with a pan-tilt system and a Kinect camera. Task completion time and accuracy were used as metrics to assess the users' performance, while the NASA-TLX questionnaire was used to evaluate the users' workload. A study with 24 participants indicated that choice of interaction modality had significant effect on completion time (F (2, 61) = 84.874, p <; 0.001), accuracy (F (2, 29) = 4.937, p = 0.016), and workload (F (2, 68) = 11.948, p <; 0.001). The direct physical interaction required less time, provided more accuracy and less workload than the two contactless interaction modalities. Between the two contactless interaction modalities, the person-following interaction modality was systematically better than the pointing-control one: The participants completed the tasks faster with less workload.

The video shows the interaction with a customized Kompa\"{i} robot. The robot consists of the Robosoft's robuLAB10 platform, tablet PC, and a Microsoft Kinect camera mounted on a pan-tilt system. A visual control algorithm provides continuous person tracking. The newly developed robot features include gesture recognition, person following, navigation with pointing, and force control, which were integrated with the Robosoft's robuBOX SDK and the Karto SLAM algorithms. The video demonstrates all the features and puts the robot in use in an everyday home scenario.

In this paper, we propose a Modified Distributed Bees Algorithm (MDBA) for multi-sensor task allocation in a supply chain security scenario. The MDBA assigns sensors to the upcoming tasks using a decentralized, probabilistic approach to maximize information gain while minimizing costs. Tasks are allocated based on sensors' performance, tasks' priorities and the mutual sensor-task distances. Simulation analysis compared different algorithms and indicated improved performance of 15% by using MDBA with respect to the second-best algorithm.

New generations of Ground based Air Defense Systems of Systems used in modernized Armed Forces maintain the architectures used in their previous versions, constrained by hierarchical radio communications and centralized Command and Control restrictions. Current state-of-the-art technology in these areas allows a refocusing of these architectures towards a swarm approach. Advantages in terms of effectiveness, scalability and survivability are identified, and a preliminary set of swarm algorithms are proposed and analyzed. Upgrades of these preliminary algorithms are proposed as future research areas.

In this paper, Ant Colony System (ACS) algorithm is applied for edge detection in grayscale images. The novelty of the proposed method is to extract a set of images from the original grayscale image using Multiscale Adaptive Gain for image contrast enhancement and then apply the ACS algorithm to detect the edges on each of the extracted images. The resulting set of images represents the pheromone trails matrices which are summed to produce the output image. The image contrast enhancement makes ACS algorithm more effective when accumulating pheromone trails on the true edge pixels. The results of the experiments are presented to confirm the effectiveness of the proposed method.

Breast cancer is one of the leading causes to women mortality in the world. Cluster of Microcalcifications (MCCs) in mammograms can be an important early sign of breast cancer, the detection is important to prevent and treat the disease. In this paper, we present a novel method for the detection of MCCs in mammograms which consists of image enhancement by histogram adaptive equalization technique, MCCs edge detection by Coordinate Logic Filters (CLF), generation, clustering and labelling of suboptimal features vectors by Self Organizing Map (SOM) Neural Network. The experiment results show that the proposed method can locate MCCs in an efficient way.