Francesca Odone

Roberto Di Via, Irina Voiculescu, Francesca Odone et al.

Anatomical landmark detection is a fundamental task in medical image analysis supporting a wide range of diagnostic and interventional workflows. Although recent methods have achieved sub-millimetric localisation, accuracy alone is not sufficient for clinical deployment, requiring reliability and robustness in prediction. Despite its clinical relevance, the impact of representation learning in this context is still underexplored. In this work, we introduce CDPM-align, a multi-scale guidance-aligned conditional diffusion pre-training for anatomical landmark detection. Our experimental setup focuses on a few images and a few annotation regimes. Specifically, we employ three popular heterogeneous small-scale benchmark datasets for representation learning via conditional generative pre-training. Furthermore, we consider low-annotation scenarios for the downstream task of landmark detection, with 10 and 25 annotated images, reflecting realistic trade-offs between clinical effort and resource constraints for annotations. Our results confirm that generative pre-training enables the model to learn a robust representation. This improves both accuracy and uncertainty on the downstream tasks, advancing towards safe and efficient clinical deployment.

Roberto Di Via, Vito Paolo Pastore, Francesca Odone et al.

Many clinical screening decisions are based on angle measurements. In particular, FemoroAcetabular Impingement (FAI) screening relies on angles traditionally measured on X-rays. However, assessing the height and span of the impingement area requires also a 3D view through an MRI scan. The two modalities inform the surgeon on different aspects of the condition. In this work, we conduct a matched-cohort validation study (89 patients, paired MRI/X-ray) using standard heatmap regression architectures to assess cross-modality clinical equivalence. Seen that landmark detection has been proven effective on X-rays, we show that MRI also achieves equivalent localisation and diagnostic accuracy for cam-type impingement. Our method demonstrates clinical feasibility for FAI assessment in coronal views of 3D MRI volumes, opening the possibility for volumetric analysis through placing further landmarks. These results support integrating automated FAI assessment into routine MRI workflows. Code is released at https://github.com/Malga-Vision/Landmarks-Hip-Conditions

Vito Paolo Pastore, Massimiliano Ciranni, Davide Marinelli et al.

It is widely recognized that deep neural networks are sensitive to bias in the data. This means that during training these models are likely to learn spurious correlations between data and labels, resulting in limited generalization abilities and low performance. In this context, model debiasing approaches can be devised aiming at reducing the model's dependency on such unwanted correlations, either leveraging the knowledge of bias information or not. In this work, we focus on the latter and more realistic scenario, showing the importance of accurately predicting the bias-conflicting and bias-aligned samples to obtain compelling performance in bias mitigation. On this ground, we propose to conceive the problem of model bias from an out-of-distribution perspective, introducing a new bias identification method based on anomaly detection. We claim that when data is mostly biased, bias-conflicting samples can be regarded as outliers with respect to the bias-aligned distribution in the feature space of a biased model, thus allowing for precisely detecting them with an anomaly detection method. Coupling the proposed bias identification approach with bias-conflicting data upsampling and augmentation in a two-step strategy, we reach state-of-the-art performance on synthetic and real benchmark datasets. Ultimately, our proposed approach shows that the data bias issue does not necessarily require complex debiasing methods, given that an accurate bias identification procedure is defined. Source code is available at https://github.com/Malga-Vision/MoDAD

Paolo Didier Alfano, Marco Rando, Marco Letizia et al.

Monitoring plankton populations in situ is fundamental to preserve the aquatic ecosystem. Plankton microorganisms are in fact susceptible of minor environmental perturbations, that can reflect into consequent morphological and dynamical modifications. Nowadays, the availability of advanced automatic or semi-automatic acquisition systems has been allowing the production of an increasingly large amount of plankton image data. The adoption of machine learning algorithms to classify such data may be affected by the significant cost of manual annotation, due to both the huge quantity of acquired data and the numerosity of plankton species. To address these challenges, we propose an efficient unsupervised learning pipeline to provide accurate classification of plankton microorganisms. We build a set of image descriptors exploiting a two-step procedure. First, a Variational Autoencoder (VAE) is trained on features extracted by a pre-trained neural network. We then use the learnt latent space as image descriptor for clustering. We compare our method with state-of-the-art unsupervised approaches, where a set of pre-defined hand-crafted features is used for clustering of plankton images. The proposed pipeline outperforms the benchmark algorithms for all the plankton datasets included in our analysis, providing better image embedding properties.

Issa Mouawad, Nikolas Brasch, Fabian Manhardt et al.

Monocular 3D object detection continues to attract attention due to the cost benefits and wider availability of RGB cameras. Despite the recent advances and the ability to acquire data at scale, annotation cost and complexity still limit the size of 3D object detection datasets in the supervised settings. Self-supervised methods, on the other hand, aim at training deep networks relying on pretext tasks or various consistency constraints. Moreover, other 3D perception tasks (such as depth estimation) have shown the benefits of temporal priors as a self-supervision signal. In this work, we argue that the temporal consistency on the level of object poses, provides an important supervision signal given the strong prior on physical motion. Specifically, we propose a self-supervised loss which uses this consistency, in addition to render-and-compare losses, to refine noisy pose predictions and derive high-quality pseudo labels. To assess the effectiveness of the proposed method, we finetune a synthetically trained monocular 3D object detection model using the pseudo-labels that we generated on real data. Evaluation on the standard KITTI3D benchmark demonstrates that our method reaches competitive performance compared to other monocular self-supervised and supervised methods.

Issa Mouawad, Francesca Odone

Object detection and tracking in videos represent essential and computationally demanding building blocks for current and future visual perception systems. In order to reduce the efficiency gap between available methods and computational requirements of real-world applications, we propose to re-think one of the most successful methods for image object detection, Faster R-CNN, and extend it to the video domain. Specifically, we extend the detection framework to learn instance-level embeddings which prove beneficial for data association and re-identification purposes. Focusing on the computational aspects of detection and tracking, our proposed method reaches a very high computational efficiency necessary for relevant applications, while still managing to compete with recent and state-of-the-art methods as shown in the experiments we conduct on standard object tracking benchmarks

Jacopo Dapueto, Nicoletta Noceti, Francesca Odone

Developing meaningful and efficient representations that separate the fundamental structure of the data generation mechanism is crucial in representation learning. However, Disentangled Representation Learning has not fully shown its potential on real images, because of correlated generative factors, their resolution and limited access to ground truth labels. Specifically on the latter, we investigate the possibility of leveraging synthetic data to learn general-purpose disentangled representations applicable to real data, discussing the effect of fine-tuning and what properties of disentanglement are preserved after the transfer. We provide an extensive empirical study to address these issues. In addition, we propose a new interpretable intervention-based metric, to measure the quality of factors encoding in the representation. Our results indicate that some level of disentanglement, transferring a representation from synthetic to real data, is possible and effective.

Paolo Didier Alfano, Vito Paolo Pastore, Lorenzo Rosasco et al.

The impressive performance of deep learning architectures is associated with a massive increase in model complexity. Millions of parameters need to be tuned, with training and inference time scaling accordingly, together with energy consumption. But is massive fine-tuning always necessary? In this paper, focusing on image classification, we consider a simple transfer learning approach exploiting pre-trained convolutional features as input for a fast-to-train kernel method. We refer to this approach as \textit{top-tuning} since only the kernel classifier is trained on the target dataset. In our study, we perform more than 3000 training processes focusing on 32 small to medium-sized target datasets, a typical situation where transfer learning is necessary. We show that the top-tuning approach provides comparable accuracy with respect to fine-tuning, with a training time between one and two orders of magnitude smaller. These results suggest that top-tuning is an effective alternative to fine-tuning in small/medium datasets, being especially useful when training time efficiency and computational resources saving are crucial.

Roberto Di Via, Francesca Odone, Vito Paolo Pastore

Deep neural networks have been extensively applied in the medical domain for various tasks, including image classification, segmentation, and landmark detection. However, their application is often hindered by data scarcity, both in terms of available annotations and images. This study introduces a novel application of denoising diffusion probabilistic models (DDPMs) to the landmark detection task, specifically addressing the challenge of limited annotated data in x-ray imaging. Our key innovation lies in leveraging DDPMs for self-supervised pre-training in landmark detection, a previously unexplored approach in this domain. This method enables accurate landmark detection with minimal annotated training data (as few as 50 images), surpassing both ImageNet supervised pre-training and traditional self-supervised techniques across three popular x-ray benchmark datasets. To our knowledge, this work represents the first application of diffusion models for self-supervised learning in landmark detection, which may offer a valuable pre-training approach in few-shot regimes, for mitigating data scarcity.

Yara AlaaEldin, Enrico Simetti, Francesca Odone

Developing a robust and effective obstacle detection and tracking system for Unmanned Surface Vehicle (USV) at marine environments is a challenging task. Research efforts have been made in this area during the past years by GRAAL lab at the university of Genova that resulted in a methodology for detecting and tracking obstacles on the image plane and, then, locating them in the 3D LiDAR point cloud. In this work, we continue on the developed system by, firstly, evaluating its performance on recently published marine datasets. Then, we integrate the different blocks of the system on ROS platform where we could test it in real-time on synchronized LiDAR and camera data collected in various marine conditions available in the MIT marine datasets. We present a thorough experimental analysis of the results obtained using two approaches; one that uses sensor fusion between the camera and LiDAR to detect and track the obstacles and the other uses only the LiDAR point cloud for the detection and tracking. In the end, we propose a hybrid approach that merges the advantages of both approaches to build an informative obstacles map of the surrounding environment to the USV.

Yara AlaaEldin, Francesca Odone

Understanding the geometric and semantic properties of the scene is crucial in autonomous navigation and particularly challenging in the case of Unmanned Aerial Vehicle (UAV) navigation. Such information may be by obtained by estimating depth and semantic segmentation maps of the surrounding environment and for their practical use in autonomous navigation, the procedure must be performed as close to real-time as possible. In this paper, we leverage monocular cameras on aerial robots to predict depth and semantic maps in low-altitude unstructured environments. We propose a joint deep-learning architecture that can perform the two tasks accurately and rapidly, and validate its effectiveness on MidAir and Aeroscapes benchmark datasets. Our joint-architecture proves to be competitive or superior to the other single and joint architecture methods while performing its task fast predicting 20.2 FPS on a single NVIDIA quadro p5000 GPU and it has a low memory footprint. All codes for training and prediction can be found on this link: https://github.com/Malga-Vision/Co-SemDepth

Massimiliano Ciranni, Vito Paolo Pastore, Roberto Di Via et al.

Deep learning model effectiveness in classification tasks is often challenged by the quality and quantity of training data whenever they are affected by strong spurious correlations between specific attributes and target labels. This results in a form of bias affecting training data, which typically leads to unrecoverable weak generalization in prediction. This paper aims at facing this problem by leveraging bias amplification with generated synthetic data: we introduce Diffusing DeBias (DDB), a novel approach acting as a plug-in for common methods of unsupervised model debiasing exploiting the inherent bias-learning tendency of diffusion models in data generation. Specifically, our approach adopts conditional diffusion models to generate synthetic bias-aligned images, which replace the original training set for learning an effective bias amplifier model that we subsequently incorporate into an end-to-end and a two-step unsupervised debiasing approach. By tackling the fundamental issue of bias-conflicting training samples memorization in learning auxiliary models, typical of this type of techniques, our proposed method beats current state-of-the-art in multiple benchmark datasets, demonstrating its potential as a versatile and effective tool for tackling bias in deep learning models. Code is available at https://github.com/Malga-Vision/DiffusingDeBias

Giorgio Cantarini, Federico Figari Tomenotti, Nicoletta Noceti et al.

In this paper we introduce a novel method to estimate the head pose of people in single images starting from a small set of head keypoints. To this purpose, we propose a regression model that exploits keypoints computed automatically by 2D pose estimation algorithms and outputs the head pose represented by yaw, pitch, and roll. Our model is simple to implement and more efficient with respect to the state of the art -- faster in inference and smaller in terms of memory occupancy -- with comparable accuracy. Our method also provides a measure of the heteroscedastic uncertainties associated with the three angles, through an appropriately designed loss function; we show there is a correlation between error and uncertainty values, thus this extra source of information may be used in subsequent computational steps. As an example application, we address social interaction analysis in images: we propose an algorithm for a quantitative estimation of the level of interaction between people, starting from their head poses and reasoning on their mutual positions. The code is available at https://github.com/cantarinigiorgio/HHP-Net.

Roberto Di Via, Matteo Santacesaria, Francesca Odone et al.

In recent years, deep learning has emerged as a promising technique for medical image analysis. However, this application domain is likely to suffer from a limited availability of large public datasets and annotations. A common solution to these challenges in deep learning is the usage of a transfer learning framework, typically with a fine-tuning protocol, where a large-scale source dataset is used to pre-train a model, further fine-tuned on the target dataset. In this paper, we present a systematic study analyzing whether the usage of small-scale in-domain x-ray image datasets may provide any improvement for landmark detection over models pre-trained on large natural image datasets only. We focus on the multi-landmark localization task for three datasets, including chest, head, and hand x-ray images. Our results show that using in-domain source datasets brings marginal or no benefit with respect to an ImageNet out-of-domain pre-training. Our findings can provide an indication for the development of robust landmark detection systems in medical images when no large annotated dataset is available.

Jacopo Dapueto, Vito Paolo Pastore, Nicoletta Noceti et al.

Microscopy image analysis is fundamental for different applications, from diagnosis to synthetic engineering and environmental monitoring. Modern acquisition systems have granted the possibility to acquire an escalating amount of images, requiring a consequent development of a large collection of deep learning-based automatic image analysis methods. Although deep neural networks have demonstrated great performance in this field, interpretability, an essential requirement for microscopy image analysis, remains an open challenge. This work proposes a Disentangled Representation Learning (DRL) methodology to enhance model interpretability for microscopy image classification. Exploiting benchmark datasets from three different microscopic image domains (plankton, yeast vacuoles, and human cells), we show how a DRL framework, based on transferring a representation learnt from synthetic data, can provide a good trade-off between accuracy and interpretability in this domain.

Issa Mouawad, Nikolas Brasch, Fabian Manhardt et al.

For autonomous vehicles, driving safely is highly dependent on the capability to correctly perceive the environment in 3D space, hence the task of 3D object detection represents a fundamental aspect of perception. While 3D sensors deliver accurate metric perception, monocular approaches enjoy cost and availability advantages that are valuable in a wide range of applications. Unfortunately, training monocular methods requires a vast amount of annotated data. Interestingly, self-supervised approaches have recently been successfully applied to ease the training process and unlock access to widely available unlabelled data. While related research leverages different priors including LIDAR scans and stereo images, such priors again limit usability. Therefore, in this work, we propose a novel approach to self-supervise 3D object detection purely from RGB sequences alone, leveraging multi-view constraints and weak labels. Our experiments on KITTI 3D dataset demonstrate performance on par with state-of-the-art self-supervised methods using LIDAR scans or stereo images.

Vipul Nair, Paul Hemeren, Alessia Vignolo et al.

Understanding which features humans rely on -- in visually recognizing action similarity is a crucial step towards a clearer picture of human action perception from a learning and developmental perspective. In the present work, we investigate to which extent a computational model based on kinematics can determine action similarity and how its performance relates to human similarity judgments of the same actions. To this aim, twelve participants perform an action similarity task, and their performances are compared to that of a computational model solving the same task. The chosen model has its roots in developmental robotics and performs action classification based on learned kinematic primitives. The comparative experiment results show that both the model and human participants can reliably identify whether two actions are the same or not. However, the model produces more false hits and has a greater selection bias than human participants. A possible reason for this is the particular sensitivity of the model towards kinematic primitives of the presented actions. In a second experiment, human participants' performance on an action identification task indicated that they relied solely on kinematic information rather than on action semantics. The results show that both the model and human performance are highly accurate in an action similarity task based on kinematic-level features, which can provide an essential basis for classifying human actions.

Philipe A. Dias, Damiano Malafronte, Henry Medeiros et al.

Effective assisted living environments must be able to perform inferences on how their occupants interact with one another as well as with surrounding objects. To accomplish this goal using a vision-based automated approach, multiple tasks such as pose estimation, object segmentation and gaze estimation must be addressed. Gaze direction in particular provides some of the strongest indications of how a person interacts with the environment. In this paper, we propose a simple neural network regressor that estimates the gaze direction of individuals in a multi-camera assisted living scenario, relying only on the relative positions of facial keypoints collected from a single pose estimation model. To handle cases of keypoint occlusion, our model exploits a novel confidence gated unit in its input layer. In addition to the gaze direction, our model also outputs an estimation of its own prediction uncertainty. Experimental results on a public benchmark demonstrate that our approach performs on pair with a complex, dataset-specific baseline, while its uncertainty predictions are highly correlated to the actual angular error of corresponding estimations. Finally, experiments on images from a real assisted living environment demonstrate the higher suitability of our model for its final application.

Gaurvi Goyal, Nicoletta Noceti, Francesca Odone et al.

Analysis and interpretation of egocentric video data is becoming more and more important with the increasing availability and use of wearable cameras. Exploring and fully understanding affinities and differences between ego and allo (or third-person) vision is paramount for the design of effective methods to process, analyse and interpret egocentric data. In addition, a deeper understanding of ego-vision and its peculiarities may enable new research perspectives in which first person viewpoints can act either as a mean for easily acquiring large amounts of data to be employed in general-purpose recognition systems, and as a challenging test-bed to assess the usability of techniques specifically tailored to deal with allocentric vision on more challenging settings. Our work, with an eye to cognitive science findings, leverages transfer learning in Convolutional Neural Networks to demonstrate capabilities and limitations of an implicitly learnt view-invariant representation in the specific case of action recognition.

Giulia Pasquale, Carlo Ciliberto, Francesca Odone et al.

We report on an extensive study of the benefits and limitations of current deep learning approaches to object recognition in robot vision scenarios, introducing a novel dataset used for our investigation. To avoid the biases in currently available datasets, we consider a natural human-robot interaction setting to design a data-acquisition protocol for visual object recognition on the iCub humanoid robot. Analyzing the performance of off-the-shelf models trained off-line on large-scale image retrieval datasets, we show the necessity for knowledge transfer. We evaluate different ways in which this last step can be done, and identify the major bottlenecks affecting robotic scenarios. By studying both object categorization and identification problems, we highlight key differences between object recognition in robotics applications and in image retrieval tasks, for which the considered deep learning approaches have been originally designed. In a nutshell, our results confirm the remarkable improvements yield by deep learning in this setting, while pointing to specific open challenges that need be addressed for seamless deployment in robotics.

Miguel A. Duval-Poo, Nicoletta Noceti, Francesca Odone et al.

Shearlets are a relatively new directional multi-scale framework for signal analysis, which have been shown effective to enhance signal discontinuities such as edges and corners at multiple scales. In this work we address the problem of detecting and describing blob-like features in the shearlets framework. We derive a measure which is very effective for blob detection and closely related to the Laplacian of Gaussian. We demonstrate the measure satisfies the perfect scale invariance property in the continuous case. In the discrete setting, we derive algorithms for blob detection and keypoint description. Finally, we provide qualitative justifications of our findings as well as a quantitative evaluation on benchmark data. We also report an experimental evidence that our method is very suitable to deal with compressed and noisy images, thanks to the sparsity property of shearlets.

Giulia Pasquale, Carlo Ciliberto, Francesca Odone et al.

The ability to visually recognize objects is a fundamental skill for robotics systems. Indeed, a large variety of tasks involving manipulation, navigation or interaction with other agents, deeply depends on the accurate understanding of the visual scene. Yet, at the time being, robots are lacking good visual perceptual systems, which often become the main bottleneck preventing the use of autonomous agents for real-world applications. Lately in computer vision, systems that learn suitable visual representations and based on multi-layer deep convolutional networks are showing remarkable performance in tasks such as large-scale visual recognition and image retrieval. To this regard, it is natural to ask whether such remarkable performance would generalize also to the robotic setting. In this paper we investigate such possibility, while taking further steps in developing a computational vision system to be embedded on a robotic platform, the iCub humanoid robot. In particular, we release a new dataset ({\sc iCubWorld28}) that we use as a benchmark to address the question: {\it how many objects can iCub recognize?} Our study is developed in a learning framework which reflects the typical visual experience of a humanoid robot like the iCub. Experiments shed interesting insights on the strength and weaknesses of current computer vision approaches applied in real robotic settings.

Stefano Piana, Alessandra Staglianò, Francesca Odone et al.

Although psychological research indicates that bodily expressions convey important affective information, to date research in emotion recognition focused mainly on facial expression or voice analysis. In this paper we propose an approach to realtime automatic emotion recognition from body movements. A set of postural, kinematic, and geometrical features are extracted from sequences 3D skeletons and fed to a multi-class SVM classifier. The proposed method has been assessed on data acquired through two different systems: a professionalgrade optical motion capture system, and Microsoft Kinect. The system has been assessed on a "six emotions" recognition problem, and using a leave-one-subject-out cross validation strategy, reached an overall recognition rate of 61.3% which is very close to the recognition rate of 61.9% obtained by human observers. To provide further testing of the system, two games were developed, where one or two users have to interact to understand and express emotions with their body.

Sean Ryan Fanello, Carlo Ciliberto, Matteo Santoro et al.

In this paper we present and start analyzing the iCub World data-set, an object recognition data-set, we acquired using a Human-Robot Interaction (HRI) scheme and the iCub humanoid robot platform. Our set up allows for rapid acquisition and annotation of data with corresponding ground truth. While more constrained in its scopes -- the iCub world is essentially a robotics research lab -- we demonstrate how the proposed data-set poses challenges to current recognition systems. The iCubWorld data-set is publicly available. The data-set can be downloaded from: http://www.iit.it/en/projects/data-sets.html.