Pasquale De Marinis
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Pasquale De Marinis has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Gennaro Vessio
Department of Computer Science, University of Bari Aldo Moro, Bari, Italy
12 shared publications - Giovanna Castellano
Department of Computer Science, University of Bari Aldo Moro, Bari, Italy
12 shared publications - Nicola Fanelli
University of Bari Aldo Moro
2 shared publications - Raffaele Scaringi
Department of Computer Science, University of Bari Aldo Moro, Bari, Italy
2 shared publications - Uzay Kaymak
Jheronimus Academy of Data Science Eindhoven University of Technology Eindhoven The Netherlands
2 shared publications - Antonio Iammarino
University of Bari Aldo Moro
1 shared publication
Latest Publications
- 1 October 2026
- Published by Springer Nature
- p. 124-138
Understanding How MLLMs Describe Artworks Using Token Activation Maps
- 1 October 2026
- Published by Springer Nature
- p. 124-138
Understanding How MLLMs Describe Artworks Using Token Activation Maps
- 28 September 2026
- Published by Springer Nature
- p. 471-485
Towards Real-Time Drone Vision for Road Safety
- 28 September 2026
- Published by Springer Nature
- p. 471-485
Towards Real-Time Drone Vision for Road Safety
- 4 August 2026
- Published by Springer Nature
- p. 545-559
WeedDiffusion: A Dual-Branch Synthetic Augmentation Framework for Weed Mapping
- 4 August 2026
- Published by Springer Nature
- p. 545-559
WeedDiffusion: A Dual-Branch Synthetic Augmentation Framework for Weed Mapping
- Open Access
- 5 June 2026
- Published by Elsevier in Pattern Recognition Letters
- Vol. 207, 47-54
Take a peek: Efficient encoder adaptation for few-shot semantic segmentation via LoRA
- 5 June 2026
- Published by Elsevier in Pattern Recognition Letters
- Vol. 207, 47-54
Open AccessTake a peek: Efficient encoder adaptation for few-shot semantic segmentation via LoRA
- Open Access
Explainable Fuzzy GNNs for Leak Detection in Water Distribution Networks
Timely leak detection in water distribution networks is critical for conserving resources and maintaining operational efficiency. Although Graph Neural Networks (GNNs) excel at capturing spatial-temporal dependencies in sensor data, their black-box nature and the limited work on graph-based explainable models for water networks hinder practical adoption. We propose an explainable GNN framework that integrates mutual information to identify critical network regions and fuzzy logic to provide clear, rule-based explanations for node classification tasks. After benchmarking several GNN architectures, we selected the generalized graph convolution network (GENConv) for its superior performance and developed a fuzzy-enhanced variant that offers intuitive explanations for classified leak locations. Our fuzzy graph neural network (FGENConv) achieved Graph F1 scores of 0.889 for detection and 0.814 for localization, slightly below the crisp GENConv 0.938 and 0.858, respectively. Yet it compensates by providing spatially localized, fuzzy rule-based explanations. By striking the right balance between precision and explainability, the proposed fuzzy network could enable hydraulic engineers to validate predicted leak locations, conserve human resources, and optimize maintenance strategies. The code is available at github.com/pasqualedem/GNNLeakDetection.
Open AccessExplainable Fuzzy GNNs for Leak Detection in Water Distribution Networks
Timely leak detection in water distribution networks is critical for conserving resources and maintaining operational efficiency. Although Graph Neural Networks (GNNs) excel at capturing spatial-temporal dependencies in sensor data, their black-box nature and the limited work on graph-based explainable models for water networks hinder practical adoption. We propose an explainable GNN framework that integrates mutual information to identify critical network regions and fuzzy logic to provide clear, rule-based explanations for node classification tasks. After benchmarking several GNN architectures, we selected the generalized graph convolution network (GENConv) for its superior performance and developed a fuzzy-enhanced variant that offers intuitive explanations for classified leak locations. Our fuzzy graph neural network (FGENConv) achieved Graph F1 scores of 0.889 for detection and 0.814 for localization, slightly below the crisp GENConv 0.938 and 0.858, respectively. Yet it compensates by providing spatially localized, fuzzy rule-based explanations. By striking the right balance between precision and explainability, the proposed fuzzy network could enable hydraulic engineers to validate predicted leak locations, conserve human resources, and optimize maintenance strategies. The code is available at github.com/pasqualedem/GNNLeakDetection.
- 2 January 2026
- Published by Springer Nature
- p. 128-139
A Weakly-Supervised Learning Approach for RGB Crop Detection Using UAV Imagery
- 2 January 2026
- Published by Springer Nature
- p. 128-139
A Weakly-Supervised Learning Approach for RGB Crop Detection Using UAV Imagery
- 2 January 2026
- Published by Springer Nature
- p. 116-127
UKANWeed: An Application of Kolmogorov-Arnold Networks to Weed Mapping
- 2 January 2026
- Published by Springer Nature
- p. 116-127
UKANWeed: An Application of Kolmogorov-Arnold Networks to Weed Mapping
- Open Access
Take a Peek: Efficient Encoder Adaptation for Few-Shot Semantic Segmentation via LoRA
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.Open AccessTake a Peek: Efficient Encoder Adaptation for Few-Shot Semantic Segmentation via LoRA
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. - Open Access
DistillFSS: Synthesizing Few-Shot Knowledge into a Lightweight Segmentation Model
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.
Open AccessDistillFSS: Synthesizing Few-Shot Knowledge into a Lightweight Segmentation Model
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.
- Open Access
- 21 October 2025
- Published by IOS Press
- p. 4016-4023
Label Anything: Multi-Class Few-Shot Semantic Segmentation with Visual Prompts
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: (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; (ii) we propose a novel architecture based on transformers and attention mechanisms; and (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-20i 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.
- 21 October 2025
- Published by IOS Press
- p. 4016-4023
Open AccessLabel Anything: Multi-Class Few-Shot Semantic Segmentation with Visual Prompts
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: (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; (ii) we propose a novel architecture based on transformers and attention mechanisms; and (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-20i 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.