Gennaro Vessio

Pasquale De Marinis, Nicola Fanelli, Raffaele Scaringi et al.

Few-shot semantic segmentation aims to segment objects from previously unseen classes using only a limited number of labeled examples. In this paper, we introduce Label Anything, a novel transformer-based architecture designed for multi-prompt, multi-way few-shot semantic segmentation. Our approach leverages diverse visual prompts -- points, bounding boxes, and masks -- to create a highly flexible and generalizable framework that significantly reduces annotation burden while maintaining high accuracy. Label Anything makes three key contributions: ($\textit{i}$) we introduce a new task formulation that relaxes conventional few-shot segmentation constraints by supporting various types of prompts, multi-class classification, and enabling multiple prompts within a single image; ($\textit{ii}$) we propose a novel architecture based on transformers and attention mechanisms; and ($\textit{iii}$) we design a versatile training procedure allowing our model to operate seamlessly across different $N$-way $K$-shot and prompt-type configurations with a single trained model. Our extensive experimental evaluation on the widely used COCO-$20^i$ benchmark demonstrates that Label Anything achieves state-of-the-art performance among existing multi-way few-shot segmentation methods, while significantly outperforming leading single-class models when evaluated in multi-class settings. Code and trained models are available at https://github.com/pasqualedem/LabelAnything.

Pasquale De Marinis, Gennaro Vessio, Giovanna Castellano

Few-shot semantic segmentation (FSS) aims to segment novel classes in query images using only a small annotated support set. While prior research has mainly focused on improving decoders, the encoder's limited ability to extract meaningful features for unseen classes remains a key bottleneck. In this work, we introduce \textit{Take a Peek} (TaP), a simple yet effective method that enhances encoder adaptability for both FSS and cross-domain FSS (CD-FSS). TaP leverages Low-Rank Adaptation (LoRA) to fine-tune the encoder on the support set with minimal computational overhead, enabling fast adaptation to novel classes while mitigating catastrophic forgetting. Our method is model-agnostic and can be seamlessly integrated into existing FSS pipelines. Extensive experiments across multiple benchmarks--including COCO $20^i$, Pascal $5^i$, and cross-domain datasets such as DeepGlobe, ISIC, and Chest X-ray--demonstrate that TaP consistently improves segmentation performance across diverse models and shot settings. Notably, TaP delivers significant gains in complex multi-class scenarios, highlighting its practical effectiveness in realistic settings. A rank sensitivity analysis also shows that strong performance can be achieved even with low-rank adaptations, ensuring computational efficiency. By addressing a critical limitation in FSS--the encoder's generalization to novel classes--TaP paves the way toward more robust, efficient, and generalizable segmentation systems. The code is available at https://github.com/pasqualedem/TakeAPeek.

Giovanna Castellano, Eugenio Cotardo, Corrado Mencar et al.

Crowd analysis from drones has attracted increasing attention in recent times due to the ease of use and affordable cost of these devices. However, how this technology can provide a solution to crowd flow detection is still an unexplored research question. To this end, we propose a crowd flow detection method for video sequences shot by a drone. The method is based on a fully-convolutional network that learns to perform crowd clustering in order to detect the centroids of crowd-dense areas and track their movement in consecutive frames. The proposed method proved effective and efficient when tested on the Crowd Counting datasets of the VisDrone challenge, characterized by video sequences rather than still images. The encouraging results show that the proposed method could open up new ways of analyzing high-level crowd behavior from drones.

Lorenzo Loconte, Antonio Mari, Gennaro Gala et al.

This paper establishes a rigorous connection between circuit representations and tensor factorizations, two seemingly distinct yet fundamentally related areas. By connecting these fields, we highlight a series of opportunities that can benefit both communities. Our work generalizes popular tensor factorizations within the circuit language, and unifies various circuit learning algorithms under a single, generalized hierarchical factorization framework. Specifically, we introduce a modular "Lego block" approach to build tensorized circuit architectures. This, in turn, allows us to systematically construct and explore various circuit and tensor factorization models while maintaining tractability. This connection not only clarifies similarities and differences in existing models, but also enables the development of a comprehensive pipeline for building and optimizing new circuit/tensor factorization architectures. We show the effectiveness of our framework through extensive empirical evaluations, and highlight new research opportunities for tensor factorizations in probabilistic modeling.

Ivan Rinaldi, Matteo Mendula, Nicola Fanelli et al.

Music generation has advanced markedly through multimodal deep learning, enabling models to synthesize audio from text and, more recently, from images. However, existing image-conditioned systems suffer from two fundamental limitations: (i) they are typically trained on natural photographs, limiting their ability to capture the richer semantic, stylistic, and cultural content of artworks; and (ii) most rely on an image-to-text conversion stage, using language as a semantic shortcut that simplifies conditioning but prevents direct visual-to-audio learning. Motivated by these gaps, we introduce ArtSound, a large-scale multimodal dataset of 105,884 artwork-music pairs enriched with dual-modality captions, obtained by extending ArtGraph and the Free Music Archive. We further propose ArtToMus, the first framework explicitly designed for direct artwork-to-music generation, which maps digitized artworks to music without image-to-text translation or language-based semantic supervision. The framework projects visual embeddings into the conditioning space of a latent diffusion model, enabling music synthesis guided solely by visual information. Experimental results show that ArtToMus generates musically coherent and stylistically consistent outputs that reflect salient visual cues of the source artworks. While absolute alignment scores remain lower than those of text-conditioned systems-as expected given the substantially increased difficulty of removing linguistic supervision-ArtToMus achieves competitive perceptual quality and meaningful cross-modal correspondence. This work establishes direct visual-to-music generation as a distinct and challenging research direction, and provides resources that support applications in multimedia art, cultural heritage, and AI-assisted creative practice. Code and dataset will be publicly released upon acceptance.

Nicola Fanelli, Gennaro Vessio, Giovanna Castellano

Inpainting focuses on filling missing or corrupted regions of an image to blend seamlessly with its surrounding content and style. While conditional diffusion models have proven effective for text-guided inpainting, we introduce the novel task of multi-mask inpainting, where multiple regions are simultaneously inpainted using distinct prompts. Furthermore, we design a fine-tuning procedure for multimodal LLMs, such as LLaVA, to generate multi-mask prompts automatically using corrupted images as inputs. These models can generate helpful and detailed prompt suggestions for filling the masked regions. The generated prompts are then fed to Stable Diffusion, which is fine-tuned for the multi-mask inpainting problem using rectified cross-attention, enforcing prompts onto their designated regions for filling. Experiments on digitized paintings from WikiArt and the Densely Captioned Images dataset demonstrate that our pipeline delivers creative and accurate inpainting results. Our code, data, and trained models are available at https://cilabuniba.github.io/i-dream-my-painting.

Nicola Fanelli, Gennaro Vessio, Giovanna Castellano

Analyzing digitized artworks presents unique challenges, requiring not only visual interpretation but also a deep understanding of rich artistic, contextual, and historical knowledge. We introduce ArtSeek, a multimodal framework for art analysis that combines multimodal large language models with retrieval-augmented generation. Unlike prior work, our pipeline relies only on image input, enabling applicability to artworks without links to Wikidata or Wikipedia-common in most digitized collections. ArtSeek integrates three key components: an intelligent multimodal retrieval module based on late interaction retrieval, a contrastive multitask classification network for predicting artist, genre, style, media, and tags, and an agentic reasoning strategy enabled through in-context examples for complex visual question answering and artwork explanation via Qwen2.5-VL. Central to this approach is WikiFragments, a Wikipedia-scale dataset of image-text fragments curated to support knowledge-grounded multimodal reasoning. Our framework achieves state-of-the-art results on multiple benchmarks, including a +8.4% F1 improvement in style classification over GraphCLIP and a +7.1 BLEU@1 gain in captioning on ArtPedia. Qualitative analyses show that ArtSeek can interpret visual motifs, infer historical context, and retrieve relevant knowledge, even for obscure works. Though focused on visual arts, our approach generalizes to other domains requiring external knowledge, supporting scalable multimodal AI research. Both the dataset and the source code will be made publicly available at https://github.com/cilabuniba/artseek.

Pasquale De Marinis, Pieter M. Blok, Uzay Kaymak et al.

Cross-Domain Few-Shot Semantic Segmentation (CD-FSS) seeks to segment unknown classes in unseen domains using only a few annotated examples. This setting is inherently challenging: source and target domains exhibit substantial distribution shifts, label spaces are disjoint, and support images are scarce--making standard episodic methods unreliable and computationally demanding at test time. To address these constraints, we propose DistillFSS, a framework that embeds support-set knowledge directly into a model's parameters through a teacher--student distillation process. By internalizing few-shot reasoning into a dedicated layer within the student network, DistillFSS eliminates the need for support images at test time, enabling fast, lightweight inference, while allowing efficient extension to novel classes in unseen domains through rapid teacher-driven specialization. Combined with fine-tuning, the approach scales efficiently to large support sets and significantly reduces computational overhead. To evaluate the framework under realistic conditions, we introduce a new CD-FSS benchmark spanning medical imaging, industrial inspection, and remote sensing, with disjoint label spaces and variable support sizes. Experiments show that DistillFSS matches or surpasses state-of-the-art baselines, particularly in multi-class and multi-shot scenarios, while offering substantial efficiency gains. The code is available at https://github.com/pasqualedem/DistillFSS.

Gennaro Vessio

Deep neural networks typically rely on the representation produced by their final hidden layer to make predictions, implicitly assuming that this single vector fully captures the semantics encoded across all preceding transformations. However, intermediate layers contain rich and complementary information -- ranging from low-level patterns to high-level abstractions -- that is often discarded when the decision head depends solely on the last representation. This paper revisits the role of the output layer and introduces LAYA (Layer-wise Attention Aggregator), a novel output head that dynamically aggregates internal representations through attention. Instead of projecting only the deepest embedding, LAYA learns input-conditioned attention weights over layer-wise features, yielding an interpretable and architecture-agnostic mechanism for synthesizing predictions. Experiments on vision and language benchmarks show that LAYA consistently matches or improves the performance of standard output heads, with relative gains of up to about one percentage point in accuracy, while providing explicit layer-attribution scores that reveal how different abstraction levels contribute to each decision. Crucially, these interpretability signals emerge directly from the model's computation, without any external post hoc explanations. The code to reproduce LAYA is publicly available at: https://github.com/gvessio/LAYA.

Pasquale De Marinis, Uzay Kaymak, Rogier Brussee et al.

Few-Shot Semantic Segmentation (FSS) models achieve strong performance in segmenting novel classes with minimal labeled examples, yet their decision-making processes remain largely opaque. While explainable AI has advanced significantly in standard computer vision tasks, interpretability in FSS remains virtually unexplored despite its critical importance for understanding model behavior and guiding support set selection in data-scarce scenarios. This paper introduces the first dedicated method for interpreting matching-based FSS models by leveraging their inherent structural properties. Our Affinity Explainer approach extracts attribution maps that highlight which pixels in support images contribute most to query segmentation predictions, using matching scores computed between support and query features at multiple feature levels. We extend standard interpretability evaluation metrics to the FSS domain and propose additional metrics to better capture the practical utility of explanations in few-shot scenarios. Comprehensive experiments on FSS benchmark datasets, using different models, demonstrate that our Affinity Explainer significantly outperforms adapted standard attribution methods. Qualitative analysis reveals that our explanations provide structured, coherent attention patterns that align with model architectures and and enable effective model diagnosis. This work establishes the foundation for interpretable FSS research, enabling better model understanding and diagnostic for more reliable few-shot segmentation systems. The source code is publicly available at https://github.com/pasqualedem/AffinityExplainer.

Moises Diaz, Miguel Angel Ferrer, Donato Impedovo et al.

Computer aided diagnosis systems can provide non-invasive, low-cost tools to support clinicians. These systems have the potential to assist the diagnosis and monitoring of neurodegenerative disorders, in particular Parkinson's disease (PD). Handwriting plays a special role in the context of PD assessment. In this paper, the discriminating power of "dynamically enhanced" static images of handwriting is investigated. The enhanced images are synthetically generated by exploiting simultaneously the static and dynamic properties of handwriting. Specifically, we propose a static representation that embeds dynamic information based on: (i) drawing the points of the samples, instead of linking them, so as to retain temporal/velocity information; and (ii) adding pen-ups for the same purpose. To evaluate the effectiveness of the new handwriting representation, a fair comparison between this approach and state-of-the-art methods based on static and dynamic handwriting is conducted on the same dataset, i.e. PaHaW. The classification workflow employs transfer learning to extract meaningful features from multiple representations of the input data. An ensemble of different classifiers is used to achieve the final predictions. Dynamically enhanced static handwriting is able to outperform the results obtained by using static and dynamic handwriting separately.

Moises Diaz, Miguel A. Ferrer, Jose Juan Quintana et al.

Online signature parameters, which are based on human characteristics, broaden the applicability of an automatic signature verifier. Although kinematic and dynamic features have previously been suggested, accurately measuring features such as arm and forearm torques remains challenging. We present two approaches for estimating angular velocities, angular positions, and force torques. The first approach involves using a physical UR5e robotic arm to reproduce a signature while capturing those parameters over time. The second method, a cost effective approach, uses a neural network to estimate the same parameters. Our findings demonstrate that a simple neural network model can extract effective parameters for signature verification. Training the neural network with the MCYT300 dataset and cross validating with other databases, namely, BiosecurID, Visual, Blind, OnOffSigDevanagari 75 and OnOffSigBengali 75 confirm the models generalization capability.

Moises Diaz, Miguel A. Ferrer, Gennaro Vessio

Signature verification is a critical task in many applications, including forensic science, legal judgments, and financial markets. However, current signature verification systems are often difficult to explain, which can limit their acceptance in these applications. In this paper, we propose a novel explainable offline automatic signature verifier (ASV) to support forensic handwriting examiners. Our ASV is based on a universal background model (UBM) constructed from offline signature images. It allows us to assign a questioned signature to the UBM and to a reference set of known signatures using simple distance measures. This makes it possible to explain the verifier's decision in a way that is understandable to non experts. We evaluated our ASV on publicly available databases and found that it achieves competitive performance with state of the art ASVs, even when challenging 1 versus 1 comparison are considered. Our results demonstrate that it is possible to develop an explainable ASV that is also competitive in terms of performance. We believe that our ASV has the potential to improve the acceptance of signature verification in critical applications such as forensic science and legal judgments.

Andrea Andrisani, Gennaro Vessio, Fabrizio Sgobba et al.

Quantum optical neurons (QONs) are emerging as promising computational units that leverage photonic interference to perform neural operations in an energy-efficient and physically grounded manner. Building on recent theoretical proposals, we introduce a family of QON architectures based on Hong-Ou-Mandel (HOM) and Mach-Zehnder (MZ) interferometers, incorporating different photon modulation strategies -- phase, amplitude, and intensity. These physical setups yield distinct pre-activation functions, which we implement as fully differentiable modules in software. We evaluate these QONs both in isolation and as building blocks of multilayer networks, training them on binary and multiclass image classification tasks using the MNIST and FashionMNIST datasets. Our experiments show that two configurations -- HOM-based amplitude modulation and MZ-based phase-shifted modulation -- achieve performance comparable to that of classical neurons in several settings, and in some cases exhibit faster or more stable convergence. In contrast, intensity-based encodings display greater sensitivity to distributional shifts and training instabilities. These results highlight the potential of QONs as efficient and scalable components for future quantum-inspired neural architectures and hybrid photonic-electronic systems.

Dawei Du, Longyin Wen, Pengfei Zhu et al.

Crowd counting on the drone platform is an interesting topic in computer vision, which brings new challenges such as small object inference, background clutter and wide viewpoint. However, there are few algorithms focusing on crowd counting on the drone-captured data due to the lack of comprehensive datasets. To this end, we collect a large-scale dataset and organize the Vision Meets Drone Crowd Counting Challenge (VisDrone-CC2020) in conjunction with the 16th European Conference on Computer Vision (ECCV 2020) to promote the developments in the related fields. The collected dataset is formed by $3,360$ images, including $2,460$ images for training, and $900$ images for testing. Specifically, we manually annotate persons with points in each video frame. There are $14$ algorithms from $15$ institutes submitted to the VisDrone-CC2020 Challenge. We provide a detailed analysis of the evaluation results and conclude the challenge. More information can be found at the website: \url{http://www.aiskyeye.com/}.

Giovanna Castellano, Gennaro Vessio

Clustering artworks is difficult for several reasons. On the one hand, recognizing meaningful patterns based on domain knowledge and visual perception is extremely hard. On the other hand, applying traditional clustering and feature reduction techniques to the highly dimensional pixel space can be ineffective. To address these issues, in this paper we propose DELIUS: a DEep learning approach to cLustering vIsUal artS. The method uses a pre-trained convolutional network to extract features and then feeds these features into a deep embedded clustering model, where the task of mapping the input data to a latent space is jointly optimized with the task of finding a set of cluster centroids in this latent space. Quantitative and qualitative experimental results show the effectiveness of the proposed method. DELIUS can be useful for several tasks related to art analysis, in particular visual link retrieval and historical knowledge discovery in painting datasets.

Giovanna Castellano, Giovanni Sansaro, Gennaro Vessio

Automatic art analysis has seen an ever-increasing interest from the pattern recognition and computer vision community. However, most of the current work is mainly based solely on digitized artwork images, sometimes supplemented with some metadata and textual comments. A knowledge graph that integrates a rich body of information about artworks, artists, painting schools, etc., in a unified structured framework can provide a valuable resource for more powerful information retrieval and knowledge discovery tools in the artistic domain. To this end, this paper presents ArtGraph: an artistic knowledge graph based on WikiArt and DBpedia. The graph, implemented in Neo4j, already provides knowledge discovery capabilities without having to train a learning system. In addition, the embeddings extracted from the graph are used to inject "contextual" knowledge into a deep learning model to improve the accuracy of artwork attribute prediction tasks.

Eufemia Lella, Gennaro Vessio

In this paper, we propose a novel method for mild cognitive impairment detection based on jointly exploiting the complex network and the neural network paradigm. In particular, the method is based on ensembling different brain structural "perspectives" with artificial neural networks. On one hand, these perspectives are obtained with complex network measures tailored to describe the altered brain connectivity. In turn, the brain reconstruction is obtained by combining diffusion-weighted imaging (DWI) data to tractography algorithms. On the other hand, artificial neural networks provide a means to learn a mapping from topological properties of the brain to the presence or absence of cognitive decline. The effectiveness of the method is studied on a well-known benchmark data set in order to evaluate if it can provide an automatic tool to support the early disease diagnosis. Also, the effects of balancing issues are investigated to further assess the reliability of the complex network approach to DWI data.

Moises Diaz, Momina Moetesum, Imran Siddiqi et al.

Parkinson's disease (PD) is commonly characterized by several motor symptoms, such as bradykinesia, akinesia, rigidity, and tremor. The analysis of patients' fine motor control, particularly handwriting, is a powerful tool to support PD assessment. Over the years, various dynamic attributes of handwriting, such as pen pressure, stroke speed, in-air time, etc., which can be captured with the help of online handwriting acquisition tools, have been evaluated for the identification of PD. Motion events, and their associated spatio-temporal properties captured in online handwriting, enable effective classification of PD patients through the identification of unique sequential patterns. This paper proposes a novel classification model based on one-dimensional convolutions and Bidirectional Gated Recurrent Units (BiGRUs) to assess the potential of sequential information of handwriting in identifying Parkinsonian symptoms. One-dimensional convolutions are applied to raw sequences as well as derived features; the resulting sequences are then fed to BiGRU layers to achieve the final classification. The proposed method outperformed state-of-the-art approaches on the PaHaW dataset and achieved competitive results on the NewHandPD dataset.

Giovanna Castellano, Gennaro Vessio

Clustering artworks is difficult for several reasons. On the one hand, recognizing meaningful patterns in accordance with domain knowledge and visual perception is extremely difficult. On the other hand, applying traditional clustering and feature reduction techniques to the highly dimensional pixel space can be ineffective. To address these issues, we propose to use a deep convolutional embedding model for digitized painting clustering, in which the task of mapping the raw input data to an abstract, latent space is jointly optimized with the task of finding a set of cluster centroids in this latent feature space. Quantitative and qualitative experimental results show the effectiveness of the proposed method. The model is also capable of outperforming other state-of-the-art deep clustering approaches to the same problem. The proposed method can be useful for several art-related tasks, in particular visual link retrieval and historical knowledge discovery in painting datasets.

Giovanna Castellano, Eufemia Lella, Gennaro Vessio

Visual arts are of inestimable importance for the cultural, historic and economic growth of our society. One of the building blocks of most analysis in visual arts is to find similarity relationships among paintings of different artists and painting schools. To help art historians better understand visual arts, this paper presents a framework for visual link retrieval and knowledge discovery in digital painting datasets. Visual link retrieval is accomplished by using a deep convolutional neural network to perform feature extraction and a fully unsupervised nearest neighbor mechanism to retrieve links among digitized paintings. Historical knowledge discovery is achieved by performing a graph analysis that makes it possible to study influences among artists. An experimental evaluation on a database collecting paintings by very popular artists shows the effectiveness of the method. The unsupervised strategy makes the method interesting especially in cases where metadata are scarce, unavailable or difficult to collect.