Andrea Cherubini

Giorgio Roffo, Carlo Biffi, Pietro Salvagnini et al.

To address overfitting and enhance model generalization in gastroenterological polyp size assessment, our study introduces Feature-Selection Gates (FSG) or Hard-Attention Gates (HAG) alongside Gradient Routing (GR) for dynamic feature selection. This technique aims to boost Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) by promoting sparse connectivity, thereby reducing overfitting and enhancing generalization. HAG achieves this through sparsification with learnable weights, serving as a regularization strategy. GR further refines this process by optimizing HAG parameters via dual forward passes, independently from the main model, to improve feature re-weighting. Our evaluation spanned multiple datasets, including CIFAR-100 for a broad impact assessment and specialized endoscopic datasets (REAL-Colon, Misawa, and SUN) focusing on polyp size estimation, covering over 200 polyps in more than 370,000 frames. The findings indicate that our HAG-enhanced networks substantially enhance performance in both binary and triclass classification tasks related to polyp sizing. Specifically, CNNs experienced an F1 Score improvement to 87.8% in binary classification, while in triclass classification, the ViT-T model reached an F1 Score of 76.5%, outperforming traditional CNNs and ViT-T models. To facilitate further research, we are releasing our codebase, which includes implementations for CNNs, multistream CNNs, ViT, and HAG-augmented variants. This resource aims to standardize the use of endoscopic datasets, providing public training-validation-testing splits for reliable and comparable research in gastroenterological polyp size estimation. The codebase is available at github.com/cosmoimd/feature-selection-gates.

Luca Parolari, Andrea Cherubini, Lamberto Ballan et al.

Automated colonoscopy reporting holds great potential for enhancing quality control and improving cost-effectiveness of colonoscopy procedures. A major challenge lies in the automated identification, tracking, and re-association (ReID) of polyps tracklets across full-procedure colonoscopy videos. This is essential for precise polyp counting and enables automated computation of key quality metrics, such as Adenoma Detection Rate (ADR) and Polyps Per Colonoscopy (PPC). However, polyp ReID is challenging due to variations in polyp appearance, frequent disappearance from the field of view, and occlusions. In this work, we leverage the REAL-Colon dataset, the first open-access dataset providing full-procedure videos, to define tasks, data splits and metrics for the problem of automatically count polyps in full-procedure videos, establishing an open-access framework. We re-implement previously proposed SimCLR-based methods for learning representations of polyp tracklets, both single-frame and multi-view, and adapt them to the polyp counting task. We then propose an Affinity Propagation-based clustering method to further improve ReID based on these learned representations, ultimately enhancing polyp counting. Our approach achieves state-of-the-art performance, with a polyp fragmentation rate of 6.30 and a false positive rate (FPR) below 5% on the REAL-Colon dataset. We release code at https://github.com/lparolari/towards-polyp-counting.

Ege Gursoy, Dana Kulić, Andrea Cherubini

In agricultural robotics, effective observation and localization of fruits present challenges due to occlusions caused by other parts of the tree, such as branches and leaves. These occlusions can result in false fruit localization or impede the robot from picking the fruit. The objective of this work is to push away branches that block the fruit's view to increase their visibility. Our setup consists of an RGB-D camera and a robot arm. First, we detect the occluded fruit in the RGB image and estimate its occluded part via a deep learning generative model in the depth space. The direction to push to clear the occlusions is determined using classic image processing techniques. We then introduce a 3D extension of the 2D Hough transform to detect straight line segments in the point cloud. This extension helps detect tree branches and identify the one mainly responsible for the occlusion. Finally, we clear the occlusion by pushing the branch with the robot arm. Our method uses a combination of deep learning for fruit appearance estimation, classic image processing for push direction determination, and 3D Hough transform for branch detection. We validate our perception methods through real data under different lighting conditions and various types of fruits (i.e. apple, lemon, orange), achieving improved visibility and successful occlusion clearance. We demonstrate the practical application of our approach through a real robot branch pushing demonstration.

Luca Parolari, Andrea Cherubini, Lamberto Ballan et al.

Automated polyp counting in colonoscopy is a crucial step toward automated procedure reporting and quality control, aiming to enhance the cost-effectiveness of colonoscopy screening. Counting polyps in a procedure involves detecting and tracking polyps, and then clustering tracklets that belong to the same polyp entity. Existing methods for polyp counting rely on self-supervised learning and primarily leverage visual appearance, neglecting temporal relationships in both tracklet feature learning and clustering stages. In this work, we introduce a paradigm shift by proposing a supervised contrastive loss that incorporates temporally-aware soft targets. Our approach captures intra-polyp variability while preserving inter-polyp discriminability, leading to more robust clustering. Additionally, we improve tracklet clustering by integrating a temporal adjacency constraint, reducing false positive re-associations between visually similar but temporally distant tracklets. We train and validate our method on publicly available datasets and evaluate its performance with a leave-one-out cross-validation strategy. Results demonstrate a 2.2x reduction in fragmentation rate compared to prior approaches. Our results highlight the importance of temporal awareness in polyp counting, establishing a new state-of-the-art. Code is available at https://github.com/lparolari/temporally-aware-polyp-counting.

Carlo Biffi, Giulio Antonelli, Sebastian Bernhofer et al.

Detection and diagnosis of colon polyps are key to preventing colorectal cancer. Recent evidence suggests that AI-based computer-aided detection (CADe) and computer-aided diagnosis (CADx) systems can enhance endoscopists' performance and boost colonoscopy effectiveness. However, most available public datasets primarily consist of still images or video clips, often at a down-sampled resolution, and do not accurately represent real-world colonoscopy procedures. We introduce the REAL-Colon (Real-world multi-center Endoscopy Annotated video Library) dataset: a compilation of 2.7M native video frames from sixty full-resolution, real-world colonoscopy recordings across multiple centers. The dataset contains 350k bounding-box annotations, each created under the supervision of expert gastroenterologists. Comprehensive patient clinical data, colonoscopy acquisition information, and polyp histopathological information are also included in each video. With its unprecedented size, quality, and heterogeneity, the REAL-Colon dataset is a unique resource for researchers and developers aiming to advance AI research in colonoscopy. Its openness and transparency facilitate rigorous and reproducible research, fostering the development and benchmarking of more accurate and reliable colonoscopy-related algorithms and models.

Carlo Biffi, Giorgio Roffo, Pietro Salvagnini et al.

Following recent advancements in computer-aided detection and diagnosis systems for colonoscopy, the automated reporting of colonoscopy procedures is set to further revolutionize clinical practice. A crucial yet underexplored aspect in the development of these systems is the creation of computer vision models capable of autonomously segmenting full-procedure colonoscopy videos into anatomical sections and procedural phases. In this work, we aim to create the first open-access dataset for this task and propose a state-of-the-art approach, benchmarked against competitive models. We annotated the publicly available REAL-Colon dataset, consisting of 2.7 million frames from 60 complete colonoscopy videos, with frame-level labels for anatomical locations and colonoscopy phases across nine categories. We then present ColonTCN, a learning-based architecture that employs custom temporal convolutional blocks designed to efficiently capture long temporal dependencies for the temporal segmentation of colonoscopy videos. We also propose a dual k-fold cross-validation evaluation protocol for this benchmark, which includes model assessment on unseen, multi-center data.ColonTCN achieves state-of-the-art performance in classification accuracy while maintaining a low parameter count when evaluated using the two proposed k-fold cross-validation settings, outperforming competitive models. We report ablation studies to provide insights into the challenges of this task and highlight the benefits of the custom temporal convolutional blocks, which enhance learning and improve model efficiency. We believe that the proposed open-access benchmark and the ColonTCN approach represent a significant advancement in the temporal segmentation of colonoscopy procedures, fostering further open-access research to address this clinical need.

Jihong Zhu, Andrea Cherubini, Claire Dune et al.

Deformable object manipulation (DOM) is an emerging research problem in robotics. The ability to manipulate deformable objects endows robots with higher autonomy and promises new applications in the industrial, services, and healthcare sectors. However, compared to rigid object manipulation, the manipulation of deformable objects is considerably more complex, and is still an open research problem. Addressing DOM challenges demand breakthroughs in almost all aspects of robotics, namely hardware design, sensing, (deformation) modeling, planning, and control. In this article, we review recent advances and highlight the main challenges when considering deformation in each sub-field. A particular focus of our paper lies in the discussions of these challenges and proposing future directions of research.

Luyin Hu, David Navarro-Alarcon, Andrea Cherubini et al.

In this paper, we introduce a new sensor-based control method that regulates (by means of robot motions) the heat transfer between a radiative source and an object of interest. This valuable sensorimotor capability is needed in many industrial, dermatology and field robot applications, and it is an essential component for creating machines with advanced thermo-motor intelligence. To this end, we derive a geometric-thermal-motor model which describes the relationship between the robot's active configuration and the produced dynamic thermal response. We then use the model to guide the design of two new thermal servoing controllers (one model-based and one adaptive), and analyze their stability with Lyapunov theory. To validate our method, we report a detailed experimental study with a robotic manipulator conducting autonomous thermal servoing tasks. To the best of the authors' knowledge, this is the first time that temperature regulation has been formulated as a motion control problem for robots.

Andrea Cherubini, David Navarro-Alarcon

The objective of this paper is to present a systematic review of existing sensor-based control methodologies for applications that involve direct interaction between humans and robots, in the form of either physical collaboration or safe coexistence. To this end, we first introduce the basic formulation of the sensor-servo problem, then present the most common approaches: vision-based, touch-based, audio-based, and distance-based control. Afterwards, we discuss and formalize the methods that integrate heterogeneous sensors at the control level. The surveyed body of literature is classified according to the type of sensor, to the way multiple measurements are combined, and to the target objectives and applications. Finally, we discuss open problems, potential applications, and future research directions.

Jihong Zhu, David Navarro-Alarcon, Robin Passama et al.

This paper proposes a unified vision-based manipulation framework using image contours of deformable/rigid objects. Instead of using human-defined cues, the robot automatically learns the features from processed vision data. Our method simultaneously generates -- from the same data -- both, visual features and the interaction matrix that relates them to the robot control inputs. Extraction of the feature vector and control commands is done online and adaptively, with little data for initialization. The method allows the robot to manipulate an object without knowing whether it is rigid or deformable. To validate our approach, we conduct numerical simulations and experiments with both deformable and rigid objects.

Osama Mazhar, Sofiane Ramdani, Andrea Cherubini

Intuitive user interfaces are indispensable to interact with the human centric smart environments. In this paper, we propose a unified framework that recognizes both static and dynamic gestures, using simple RGB vision (without depth sensing). This feature makes it suitable for inexpensive human-robot interaction in social or industrial settings. We employ a pose-driven spatial attention strategy, which guides our proposed Static and Dynamic gestures Network - StaDNet. From the image of the human upper body, we estimate his/her depth, along with the region-of-interest around his/her hands. The Convolutional Neural Network in StaDNet is fine-tuned on a background-substituted hand gestures dataset. It is utilized to detect 10 static gestures for each hand as well as to obtain the hand image-embeddings. These are subsequently fused with the augmented pose vector and then passed to the stacked Long Short-Term Memory blocks. Thus, human-centred frame-wise information from the augmented pose vector and from the left/right hands image-embeddings are aggregated in time to predict the dynamic gestures of the performing person. In a number of experiments, we show that the proposed approach surpasses the state-of-the-art results on the large-scale Chalearn 2016 dataset. Moreover, we transfer the knowledge learned through the proposed methodology to the Praxis gestures dataset, and the obtained results also outscore the state-of-the-art on this dataset.

David Navarro-Alarcon, Jiaming Qi, Jihong Zhu et al.

In this article, we present a new scheme that approximates unknown sensorimotor models of robots by using feedback signals only. The formulation of the uncalibrated sensor-based regulation problem is first formulated, then, we develop a computational method that distributes the model estimation problem amongst multiple adaptive units that specialise in a local sensorimotor map. Different from traditional estimation algorithms, the proposed method requires little data to train and constrain it (the number of required data points can be analytically determined) and has rigorous stability properties (the conditions to satisfy Lyapunov stability are derived). Numerical simulations and experimental results are presented to validate the proposed method.

Andrea Cherubini, Valerio Ortenzi, Akansel Cosgun et al.

We address the problem of shaping deformable plastic materials using non-prehensile actions. Shaping plastic objects is challenging, since they are difficult to model and to track visually. We study this problem, by using kinetic sand, a plastic toy material which mimics the physical properties of wet sand. Inspired by a pilot study where humans shape kinetic sand, we define two types of actions: \textit{pushing} the material from the sides and \textit{tapping} from above. The chosen actions are executed with a robotic arm using image-based visual servoing. From the current and desired view of the material, we define states based on visual features such as the outer contour shape and the pixel luminosity values. These are mapped to actions, which are repeated iteratively to reduce the image error until convergence is reached. For pushing, we propose three methods for mapping the visual state to an action. These include heuristic methods and a neural network, trained from human actions. We show that it is possible to obtain simple shapes with the kinetic sand, without explicitly modeling the material. Our approach is limited in the types of shapes it can achieve. A richer set of action types and multi-step reasoning is needed to achieve more sophisticated shapes.

David Navarro-Alarcon, Andrea Cherubini, Xiang Li

In this article, we address the problem of computing adaptive sensorimotor models that can be used for guiding the motion of robotic systems with uncertain action-to-perception relations. The formulation of the uncalibrated sensor-based control problem is first presented, then, various computational methods for building adaptive sensorimotor models are derived and analysed. The proposed methodology is exemplified with two cases of study: (i) shape control of deformable objects with unknown properties, and (ii) soft manipulation of ultrasonic probes with uncalibrated sensors.