DistriMuSe — Safe Interaction with Robots
Zoned VAE-GAN anomaly detection for interpretable industrial safety monitoring.
Posts by Collection
portfolio
Portfolio item number 1
Short description of portfolio item number 1
projects
Applied and Freelance Research Projects Applied AI
Applied research and freelance projects in medical imaging, computer vision, machine learning, and intelligent decision systems.
LIME Stratified Sampling Explainable AI
An improved LIME sampling strategy for generating more stable and reliable image explanations.
Explainable Anomaly Detection Explainable AI
A case study on building trust in anomaly detection systems using VAE-GAN models and explainable AI.
AI on Edge Devices Research/Edge AI
Lightweight AI deployment on edge devices, including Raspberry Pi-based computer vision systems.
ShapBPT: Image Feature Attributions using Data-Aware Binary Partition Trees Research
A data-aware XAI method for image feature attribution using Binary Partition Trees and hierarchical Shapley values.
ADVIS-UniGra: RGB Anomaly Detection for Safe Human–Robot Collaboration EU Projects 🇪🇺
RGB-based anomaly detection application for safety monitoring in collaborative robotics environments using synthetic industrial data.
ADVIS-SR: Real-Time Anomaly Detection for Safe Human–Robot Interaction in real industrial scenario EU Project 🇪🇺
Industrial Safety Detector using deep Generative models based Anomaly Detection on real world
publications
Object Detection and Classification: A Joint Selection and Fusion Strategy of Deep Convolutional Neural Network and SIFT Point Features journals
Published in Multimedia Tools and Applications (Springer), 2018
Object detection and classification using hybrid fusion of CNN and SIFT features.
Deep CNN and Geometric Features-Based Gastrointestinal Tract Diseases Detection and Classification from Wireless Capsule Endoscopy Images journals
Published in Journal of Experimental & Theoretical Artificial Intelligence (Taylor & Francis), 2019
Gastrointestinal disease detection and classification from WCE images using deep CNN and geometric feature fusion.
Object Detection and Classification Based on Feature Fusion and Deep Convolutional Neural Network thesis
Published in COMSATS University Islamabad, Wah Campus, 2019
Master’s thesis on object detection and classification using deep convolutional neural networks, SIFT point features, feature fusion, and entropy-based feature selection.
Classification of Gastrointestinal Diseases of Stomach from WCE Using Improved Saliency-Based Method and Discriminant Feature Selection journals
Published in Multimedia Tools and Applications (Springer), 2019
Gastrointestinal disease classification from wireless capsule endoscopy images using saliency-based segmentation and discriminant feature selection.
Region-Based Active Contour JSEG Fusion Technique for Skin Lesion Segmentation from Dermoscopic Images journals
Published in Biomedical Research, 2019
Skin lesion segmentation using region-based active contour and JSEG fusion techniques.
An Integrated Framework of Skin Lesion Detection and Recognition through Saliency Method and Optimal Deep Neural Network Feature Selection journals
Published in Neural Computing and Applications (Springer), 2019
Skin lesion detection and classification using saliency-based segmentation and optimal deep feature selection.
A Sustainable Deep Learning Framework for Object Recognition Using Multi-Layers Deep Features Fusion and Selection journals
Published in Sustainability (MDPI), 2020
A deep learning framework for object recognition using multi-layer feature fusion and feature selection.
An Optimized Approach for Breast Cancer Classification for Histopathological Images Based on Hybrid Feature Set journals
Published in Current Medical Imaging (Bentham Science), 2021
Breast cancer classification using hybrid feature extraction from histopathological images.
A Novel Light U-Net Model for Left Ventricle Segmentation Using MRI journals
Published in Mathematics (MDPI), 2023
Lightweight U-Net model for left ventricle segmentation from MRI images.
Using Stratified Sampling to Improve LIME Image Explanations conferences
Published in AAAI-24 - 38th AAAI Conference on Artificial Intelligence, 2024
A improved version of LIME to generate meaninful explanations when LIME degenerates meaningless explanations.
Can I trust my anomaly detection system? A case study based on explainable AI conferences
Published in xAI 2024 | Explainable Artificial Intelligence, 2024
A Case study to highlight use of VAE-GAN based Gen-AI approach to detect Anomalies in Industrial Inspection systems.
Acute Lymphoblastic Leukemia Classification: Deep Learning Techniques for Blood Diseases Diagnosis journals
Published in Journal of Computing & Biomedical Informatics, 2025
Deep learning-based classification of Acute Lymphoblastic Leukemia from peripheral blood smear images using EfficientNetB3.
ShapBPT: Image Feature Attributions using Data-Aware Binary Partition Trees conferences
Published in AAAI-26 | 40th Annual AAAI Conference on Artificial Intelligence, 2026
A data-aware XAI method for image feature attribution using Binary Partition Trees and hierarchical Shapley values.
Improving Trust in Safety-Critical AI Systems: Explainable AI and Anomaly Detection Frameworks for Human Safety in Smart Industries thesis
Published in University of Torino, 2026
PhD thesis on explainable AI, Shapley-based visual explanations, and anomaly detection frameworks for safety-critical industrial systems.
ShapBPT in Perspective: A Consolidated Review and an eXplainable Anomaly Detection Case Study conferences
Published in QualITA Workshop | ICPE 2026 (ACM), 2026
A consolidated review of ShapBPT and its application to explainable anomaly detection.
talks
Using Stratified Sampling to Improve LIME Image Explanations
We investigate the use of a stratified sampling approach for LIME Image, a popular model-agnostic explainable AI method for computer vision tasks, in order to reduce the artifacts generated by typical Monte Carlo samp...
Can I Trust My Anomaly Detection System? A Case Study Based on Explainable AI
A case study on the trustworthiness of VAE-based anomaly detection systems using Explainable AI, highlighting how high anomaly scores can be driven by misleading visual features.
Unexpected Condition Detector for Industrial Safety using Deep Generative Models
A consortium presentation on VAE-GAN based unexpected condition detection for industrial safety monitoring in robotic environments, developed within UC3 of the DistriMuSe EU project.
ShapBPT: Image Feature Attributions using Data-Aware Binary Partition Trees
Introduces ShapBPT, a data-aware hierarchical Shapley-based method for image feature attribution that improves efficiency, structural alignment, and interpretability in computer vision models.
Improving Trust in Safety-Critical AI Systems: Explainable AI and Anomaly Detection Frameworks for Human Safety in Smart Industries
PhD thesis defense presenting explainable AI and anomaly detection frameworks for trustworthy human safety monitoring in industrial environments.
ShapBPT in Perspective: A Consolidated Review and an eXplainable Anomaly Detection Case Study
This talk presents ShapBPT in Perspective, a consolidated review and practical case study of ShapBPT for eXplainable Anomaly Detection (XAD).
teaching
Teaching experience 1
This is a description of a teaching experience. You can use markdown like any other post.
Teaching experience 2
This is a description of a teaching experience. You can use markdown like any other post.