no code implementations • 17 May 2023 • Michele Colombo, Giacomo Boracchi, Simone Melzi
Shape matching is a fundamental problem in computer graphics with many applications.
no code implementations • ICCV 2023 • Andrea Porfiri Dal Cin, Giacomo Boracchi, Luca Magri
Traditional and deep Structure-from-Motion (SfM) methods typically operate under the assumption that the scene is rigid, i. e., the environment is static or consists of a single moving object.
1 code implementation • 16 Oct 2022 • Diego Stucchi, Luca Frittoli, Giacomo Boracchi
We introduce Class Distribution Monitoring (CDM), an effective concept-drift detection scheme that monitors the class-conditional distributions of a datastream.
1 code implementation • 23 Sep 2022 • Alberto Floris, Luca Frittoli, Diego Carrera, Giacomo Boracchi
Moreover, our CompositeNets substantially outperform existing solutions for anomaly detection on point clouds.
no code implementations • 30 Aug 2022 • Luca Frittoli, Diego Carrera, Beatrice Rossi, Pasqualina Fragneto, Giacomo Boracchi
Here we address WDM monitoring as an open-set recognition problem to accurately classify WDM in known categories and promptly detect novel patterns.
1 code implementation • 30 Aug 2022 • Andrea Bionda, Luca Frittoli, Giacomo Boracchi
Detecting anomalous regions in images is a frequently encountered problem in industrial monitoring.
1 code implementation • 30 Aug 2022 • Luca Frittoli, Diego Carrera, Giacomo Boracchi
Like many change-detection algorithms, QT-EWMA builds a model of the data distribution, in our case a QuantTree histogram, from a stationary training set.
1 code implementation • 21 Apr 2022 • Loris Giulivi, Mark James Carman, Giacomo Boracchi
Results of our user study demonstrate that humans can better understand and predict the system's decisions when perception visualizations are available, thus easing the debugging and deployment of deep models as trusted systems.
1 code implementation • 20 Apr 2022 • Loris Giulivi, Malhar Jere, Loris Rossi, Farinaz Koushanfar, Gabriela Ciocarlie, Briland Hitaj, Giacomo Boracchi
We present Adversarial Scratches: a novel L0 black-box attack, which takes the form of scratches in images, and which possesses much greater deployability than other state-of-the-art attacks.
no code implementations • CVPR 2021 • Luca Magri, Filippo Leveni, Giacomo Boracchi
We address the problem of recovering multiple structures of different classes in a dataset contaminated by noise and outliers.
no code implementations • ICCV 2021 • Andrea Porfiri Dal Cin, Luca Magri, Federica Arrigoni, Andrea Fusiello, Giacomo Boracchi
MultiSynch is a general solution that can cope with any linear group and we show to be profitably usable both on synthetic and real problems.
1 code implementation • 5 Dec 2019 • Malhar Jere, Loris Rossi, Briland Hitaj, Gabriela Ciocarlie, Giacomo Boracchi, Farinaz Koushanfar
We study black-box adversarial attacks for image classifiers in a constrained threat model, where adversaries can only modify a small fraction of pixels in the form of scratches on an image.
no code implementations • ICML 2018 • Giacomo Boracchi, Diego Carrera, Cristiano Cervellera, Danilo Macciò
Our experiments show that the proposed histograms are very effective in detecting changes in high dimensional data streams, and that the resulting thresholds can effectively control the false positive rate, even when the number of training samples is relatively small.
no code implementations • 16 Oct 2015 • Cesare Alippi, Giacomo Boracchi, Diego Carrera, Manuel Roveri
We address the problem of detecting changes in multivariate datastreams, and we investigate the intrinsic difficulty that change-detection methods have to face when the data dimension scales.