Explore the vast range of titles available.

A guiding robot aims to effectively bring people to and from specific places within environments that are possibly unknown to them. During this operation the robot should be able to detect and track the accompanied person, trying never to lose sight of her/him. A solution to minimize this event is to use an omnidirectional camera: its 360{\\deg} Field of View (FoV) guarantees that any framed object cannot leave the FoV if not occluded or very far from the sensor. However, the acquired panoramic videos introduce new challenges in perception tasks such as people detection and tracking, including the large size of the images to be processed, the distortion effects introduced by the cylindrical projection and the periodic nature of panoramic images. In this paper, we propose a set of targeted methods that allow to effectively adapt to panoramic videos a standard people detection and tracking pipeline originally designed for perspective cameras. Our methods have been implemented and tested inside a deep learning-based people detection and tracking framework with a commercial 360{\\deg} camera. Experiments performed on datasets specifically acquired for guiding robot applications and on a real service robot show the effectiveness of the proposed approach over other state-of-the-art systems. We release with this paper the acquired and annotated datasets and the open-source implementation of our method.

The ability of a robot to pick an object, known as robot grasping, is crucial for several applications, such as assembly or sorting. In such tasks, selecting the right target to pick is as essential as inferring a correct configuration of the gripper. A common solution to this problem relies on semantic segmentation models, which often show poor generalization to unseen objects and require considerable time and massive data to be trained. To reduce the need for large datasets, some grasping pipelines exploit few-shot semantic segmentation models, which are capable of recognizing new classes given a few examples. However, this often comes at the cost of limited performance and fine-tuning is required to be effective in robot grasping scenarios. In this work, we propose to overcome all these limitations by combining the impressive generalization capability reached by foundation models with a high-performing few-shot classifier, working as a score function to select the segmentation that is closer to the support set. The proposed model is designed to be embedded in a grasp synthesis pipeline. The extensive experiments using one or five examples show that our novel approach overcomes existing performance limitations, improving the state of the art both in few-shot semantic segmentation on the Graspnet-1B (+10.5% mIoU) and Ocid-grasp (+1.6% AP) datasets, and real-world few-shot grasp synthesis (+21.7% grasp accuracy). The project page is available at: https://leobarcellona.github.io/showandgrasp.github.io/

Robotic waste sorting poses significant challenges in both perception and manipulation, given the extreme variability of objects that should be recognized on a cluttered conveyor belt. While deep learning has proven effective in solving complex tasks, the necessity for extensive data collection and labeling limits its applicability in real-world scenarios like waste sorting. To tackle this issue, we introduce a data augmentation method based on a novel GAN architecture called wasteGAN. The proposed method allows to increase the performance of semantic segmentation models, starting from a very limited bunch of labeled examples, such as few as 100. The key innovations of wasteGAN include a novel loss function, a novel activation function, and a larger generator block. Overall, such innovations helps the network to learn from limited number of examples and synthesize data that better mirrors real-world distributions. We then leverage the higher-quality segmentation masks predicted from models trained on the wasteGAN synthetic data to compute semantic-aware grasp poses, enabling a robotic arm to effectively recognizing contaminants and separating waste in a real-world scenario. Through comprehensive evaluation encompassing dataset-based assessments and real-world experiments, our methodology demonstrated promising potential for robotic waste sorting, yielding performance gains of up to 5.8\\% in picking contaminants. The project page is available at https://github.com/bach05/wasteGAN.git

Objective The presence of micropapillary and solid adenocarcinoma patterns leads to a worse survival and a significantly higher tendency to recur. This study aims to assess the impact of pT descriptor combined with the presence of high-grade components on long-term outcomes in early-stage lung adenocarcinomas. Methods We retrospectively collected data of consecutive resected pT1-T3N0 lung adenocarcinoma from nine European Thoracic Centers. All patients who underwent a radical resection with lymph-node dissection between 2014 and 2017 were included. Differences in Overall Survival (OS) and Disease-Free Survival (DFS) and possible prognostic factors associated with outcomes were evaluated also after performing a propensity score matching to compare tumors containing non-high-grade and high-grade patterns. Results Among 607 patients, the majority were male and received a lobectomy. At least one high-grade histological pattern was seen in 230 cases (37.9%), of which 169 solid and 75 micropapillary. T1a-b-c without high-grade pattern had a significant better prognosis compared to T1a-b-c with high-grade pattern ( p  = 0.020), but the latter had similar OS compared to T2a ( p  = 0.277). Concurrently, T1a-b-c without micropapillary or solid patterns had a significantly better DFS compared to those with high-grade patterns ( p  = 0.034), and it was similar to T2a ( p  = 0.839). Multivariable analysis confirms the role of T descriptor according to high-grade pattern both for OS ( p  = 0.024; HR 1.285 95% CI 1.033–1.599) and DFS ( p  = 0.003; HR 1.196, 95% CI 1.054–1.344, respectively). These results were confirmed after the propensity score matching analysis. Conclusions pT1 lung adenocarcinomas with a high-grade component have similar prognosis of pT2a tumors.

OBJECTIVES Lung cancer is increasingly diagnosed as a second cancer. Our goal was to analyse the characteristics and outcomes of early-stage resected lung adenocarcinomas in patients with previous cancers (PC) and correlations with adenocarcinoma subtypes. METHODS We retrospectively reviewed data of patients radically operated on for stage I–II lung adenocarcinoma in 9 thoracic surgery departments between 2014 and 2017. Overall survival (OS) and time to disease relapse were evaluated between subgroups. RESULTS We included 700 consecutive patients. PC were present in 260 (37.1%). Breast adenocarcinoma, lung cancer and prostate cancer were the most frequent (21.5%, 11.5% and 11.2%, respectively). No significant differences in OS were observed between the PC and non-PC groups (P = 0.378), with 31 and 75 deaths, respectively. Patients with PC had smaller tumours and were more likely to receive sublobar resection and to be operated on with a minimally invasive approach. Previous gastric cancer (P = 0.042) and synchronous PC (when diagnosed up to 6 months before lung adenocarcinoma; P = 0.044) were related, with a worse OS. Colon and breast adenocarcinomas and melanomas were significantly related to a lower incidence of high grade (solid or micropapillary, P = 0.0039, P = 0.005 and P = 0.028 respectively), whereas patients affected by a previous lymphoma had a higher incidence of a micropapillary pattern (P = 0.008). CONCLUSIONS In patients with PC, we found smaller tumours more frequently treated with minimally invasive techniques and sublobar resection, probably due to a more careful follow-up. The impact on survival is not uniform and predictable; however, breast and colon cancers and melanoma showed a lower incidence of solid or micropapillary patterns whereas patients with lymphomas had a higher incidence of a micropapillary pattern.