no code implementations • 13 Feb 2025 • Francesco Ballerini, Pierluigi Zama Ramirez, Samuele Salti, Luigi Di Stefano
In this paper, we present the first framework that can ingest NeRFs with multiple architectures and perform inference on architectures unseen at training time.
no code implementations • 4 Jul 2024 • Alex Costanzino, Pierluigi Zama Ramirez, Giuseppe Lisanti, Luigi Di Stefano
Hence, in this work, we introduce a novel benchmark that evaluates methods on the original, high-resolution image and ground-truth masks, focusing on segmentation performance as a function of the size of anomalies.
no code implementations • 17 Jun 2024 • Andrea Amaduzzi, Pierluigi Zama Ramirez, Giuseppe Lisanti, Samuele Salti, Luigi Di Stefano
Multimodal Large Language Models (MLLMs) have demonstrated an excellent understanding of images and 3D data.
1 code implementation • 11 Apr 2024 • Francesco Ballerini, Pierluigi Zama Ramirez, Roberto Mirabella, Samuele Salti, Luigi Di Stefano
Neural Radiance Fields (NeRFs) have emerged as a standard framework for representing 3D scenes and objects, introducing a novel data type for information exchange and storage.
no code implementations • 4 Apr 2024 • Alex Costanzino, Pierluigi Zama Ramirez, Mirko Del Moro, Agostino Aiezzo, Giuseppe Lisanti, Samuele Salti, Luigi Di Stefano
Anomaly Detection and Segmentation (AD&S) is crucial for industrial quality control.
no code implementations • 20 Dec 2023 • Pierluigi Zama Ramirez, Luca De Luigi, Daniele Sirocchi, Adriano Cardace, Riccardo Spezialetti, Francesco Ballerini, Samuele Salti, Luigi Di Stefano
In recent years, Neural Fields (NFs) have emerged as an effective tool for encoding diverse continuous signals such as images, videos, audio, and 3D shapes.
no code implementations • CVPR 2024 • Alex Costanzino, Pierluigi Zama Ramirez, Giuseppe Lisanti, Luigi Di Stefano
The paper explores the industrial multimodal Anomaly Detection (AD) task, which exploits point clouds and RGB images to localize anomalies.
1 code implementation • 2 Oct 2023 • Adriano Cardace, Pierluigi Zama Ramirez, Francesco Ballerini, Allan Zhou, Samuele Salti, Luigi Di Stefano
While processing a field with the same reconstruction quality, we achieve task performance far superior to frameworks that process large MLPs and, for the first time, almost on par with architectures handling explicit representations.
no code implementations • 14 Sep 2023 • Andrea Amaduzzi, Giuseppe Lisanti, Samuele Salti, Luigi Di Stefano
The refined dataset, the new metric and a set of text-shape pairs validated by the user study comprise a novel, fine-grained benchmark that we publicly release to foster research on text-to-shape coherence of text-conditioned 3D generative models.
1 code implementation • ICCV 2023 • Alex Costanzino, Pierluigi Zama Ramirez, Matteo Poggi, Fabio Tosi, Stefano Mattoccia, Luigi Di Stefano
Inferring the depth of transparent or mirror (ToM) surfaces represents a hard challenge for either sensors, algorithms, or deep networks.
no code implementations • CVPR 2023 • Marco Toschi, Riccardo De Matteo, Riccardo Spezialetti, Daniele De Gregorio, Luigi Di Stefano, Samuele Salti
By leveraging the dataset, we perform an ablation study on the relighting capability of variants of the vanilla NeRF architecture and identify a lightweight architecture that can render novel views of an object under novel light conditions, which we use to establish a non-trivial baseline for the dataset.
1 code implementation • 6 Apr 2023 • Adriano Cardace, Pierluigi Zama Ramirez, Samuele Salti, Luigi Di Stefano
3D semantic segmentation is a critical task in many real-world applications, such as autonomous driving, robotics, and mixed reality.
no code implementations • 10 Feb 2023 • Luca De Luigi, Adriano Cardace, Riccardo Spezialetti, Pierluigi Zama Ramirez, Samuele Salti, Luigi Di Stefano
Implicit Neural Representations (INRs) have emerged in the last few years as a powerful tool to encode continuously a variety of different signals like images, videos, audio and 3D shapes.
no code implementations • 26 Jan 2023 • Pierluigi Zama Ramirez, Adriano Cardace, Luca De Luigi, Alessio Tonioni, Samuele Salti, Luigi Di Stefano
Besides, we propose a set of strategies to constrain the learned feature spaces, to ease learning and increase the generalization capability of the mapping network, thereby considerably improving the final performance of our framework.
no code implementations • 19 Jan 2023 • Pierluigi Zama Ramirez, Alex Costanzino, Fabio Tosi, Matteo Poggi, Samuele Salti, Stefano Mattoccia, Luigi Di Stefano
Estimating depth from images nowadays yields outstanding results, both in terms of in-domain accuracy and generalization.
no code implementations • 24 Nov 2022 • Luca De Luigi, Damiano Bolognini, Federico Domeniconi, Daniele De Gregorio, Matteo Poggi, Luigi Di Stefano
In this paper, we propose the first-ever real benchmark thought for evaluating Neural Radiance Fields (NeRFs) and, in general, Neural Rendering (NR) frameworks.
no code implementations • 15 Oct 2022 • Adriano Cardace, Riccardo Spezialetti, Pierluigi Zama Ramirez, Samuele Salti, Luigi Di Stefano
In contrast, in this work, we focus on obtaining a discriminative feature space for the target domain enforcing consistency between a point cloud and its augmented version.
no code implementations • 1 Sep 2022 • Matteo Poggi, Pierluigi Zama Ramirez, Fabio Tosi, Samuele Salti, Stefano Mattoccia, Luigi Di Stefano
We propose X-NeRF, a novel method to learn a Cross-Spectral scene representation given images captured from cameras with different light spectrum sensitivity, based on the Neural Radiance Fields formulation.
no code implementations • CVPR 2022 • Fabio Tosi, Pierluigi Zama Ramirez, Matteo Poggi, Samuele Salti, Stefano Mattoccia, Luigi Di Stefano
We address the problem of registering synchronized color (RGB) and multi-spectral (MS) images featuring very different resolution by solving stereo matching correspondences.
1 code implementation • 10 Jun 2022 • Gianluca Berardi, Luca De Luigi, Samuele Salti, Luigi Di Stefano
In particular, we show that it is possible to use representation learning to learn a fixed-size, low-dimensional embedding space of trained deep models and that such space can be explored by interpolation or optimization to attain ready-to-use models.
no code implementations • CVPR 2022 • Pierluigi Zama Ramirez, Fabio Tosi, Matteo Poggi, Samuele Salti, Stefano Mattoccia, Luigi Di Stefano
We present a novel high-resolution and challenging stereo dataset framing indoor scenes annotated with dense and accurate ground-truth disparities.
1 code implementation • 28 Oct 2021 • Filippo Aleotti, Fabio Tosi, Pierluigi Zama Ramirez, Matteo Poggi, Samuele Salti, Stefano Mattoccia, Luigi Di Stefano
We introduce a novel architecture for neural disparity refinement aimed at facilitating deployment of 3D computer vision on cheap and widespread consumer devices, such as mobile phones.
1 code implementation • 21 Oct 2021 • Adriano Cardace, Riccardo Spezialetti, Pierluigi Zama Ramirez, Samuele Salti, Luigi Di Stefano
Unsupervised Domain Adaptation (UDA) for point cloud classification is an emerging research problem with relevant practical motivations.
1 code implementation • 13 Oct 2021 • Adriano Cardace, Luca De Luigi, Pierluigi Zama Ramirez, Samuele Salti, Luigi Di Stefano
We further rely on depth to generate a large and varied set of samples to Self-Train the final model.
1 code implementation • 6 Oct 2021 • Adriano Cardace, Pierluigi Zama Ramirez, Samuele Salti, Luigi Di Stefano
Although deep neural networks have achieved remarkable results for the task of semantic segmentation, they usually fail to generalize towards new domains, especially when performing synthetic-to-real adaptation.
1 code implementation • 24 Dec 2020 • Daniele De Gregorio, Riccardo Zanella, Gianluca Palli, Luigi Di Stefano
In this paper we investigate how to effectively deploy deep learning in practical industrial settings, such as robotic grasping applications.
no code implementations • 4 Dec 2020 • Marco Boschi, Luigi Di Stefano, Martino Alessandrini
The framework is specialized here in the context of a machine vision system for automated product inspection.
1 code implementation • NeurIPS 2020 • Riccardo Spezialetti, Federico Stella, Marlon Marcon, Luciano Silva, Samuele Salti, Luigi Di Stefano
In this work, we show the feasibility of learning a robust canonical orientation for surfaces represented as point clouds.
1 code implementation • 10 Jul 2020 • Matteo Poggi, Alessio Tonioni, Fabio Tosi, Stefano Mattoccia, Luigi Di Stefano
Thus, our network architecture and adaptation algorithms realize the first real-time self-adaptive deep stereo system and pave the way for a new paradigm that can facilitate practical deployment of end-to-end architectures for dense disparity regression.
1 code implementation • CVPR 2020 • Fabio Tosi, Filippo Aleotti, Pierluigi Zama Ramirez, Matteo Poggi, Samuele Salti, Luigi Di Stefano, Stefano Mattoccia
Whole understanding of the surroundings is paramount to autonomous systems.
no code implementations • 10 Mar 2020 • Alessandro Berlati, Oliver Scheel, Luigi Di Stefano, Federico Tombari
Ambiguity is inherently present in many machine learning tasks, but especially for sequential models seldom accounted for, as most only output a single prediction.
no code implementations • 6 Nov 2019 • Marlon Marcon, Riccardo Spezialetti, Samuele Salti, Luciano Silva, Luigi Di Stefano
Object recognition in 3D point clouds is a challenging task, mainly when time is an important factor to deal with, such as in industrial applications.
1 code implementation • 11 Oct 2019 • Pierluigi Zama Ramirez, Claudio Paternesi, Luca De Luigi, Luigi Lella, Daniele De Gregorio, Luigi Di Stefano
Availability of a few, large-size, annotated datasets, like ImageNet, Pascal VOC and COCO, has lead deep learning to revolutionize computer vision research by achieving astonishing results in several vision tasks. We argue that new tools to facilitate generation of annotated datasets may help spreading data-driven AI throughout applications and domains.
no code implementations • ICCV 2019 • Riccardo Spezialetti, Samuele Salti, Luigi Di Stefano
Establishing correspondences between 3D shapes is a fundamental task in 3D Computer Vision, typically addressed by matching local descriptors.
no code implementations • 15 Sep 2019 • Riccardo Spezialetti, Samuele Salti, Luigi Di Stefano
Matching surfaces is a challenging 3D Computer Vision problem typically addressed by local features.
1 code implementation • 9 Sep 2019 • Alessio Tonioni, Matteo Poggi, Stefano Mattoccia, Luigi Di Stefano
Extensive experimental results based on standard datasets and evaluation protocols prove that our technique can address effectively the domain shift issue with both stereo and monocular depth prediction architectures and outperforms other state-of-the-art unsupervised loss functions that may be alternatively deployed to pursue domain adaptation.
1 code implementation • 5 Aug 2019 • Daniele De Gregorio, Alessio Tonioni, Gianluca Palli, Luigi Di Stefano
In this paper, we propose Augmented Reality Semi-automatic labeling (ARS), a semi-automatic method which leverages on moving a 2D camera by means of a robot, proving precise camera tracking, and an augmented reality pen to define initial object bounding box, to create large labeled datasets with minimal human intervention.
no code implementations • 17 Jul 2019 • Oscar Rahnama, Tommaso Cavallari, Stuart Golodetz, Alessio Tonioni, Thomas Joy, Luigi Di Stefano, Simon Walker, Philip H. S. Torr
Obtaining highly accurate depth from stereo images in real time has many applications across computer vision and robotics, but in some contexts, upper bounds on power consumption constrain the feasible hardware to embedded platforms such as FPGAs.
2 code implementations • ICCV 2019 • Pierluigi Zama Ramirez, Alessio Tonioni, Samuele Salti, Luigi Di Stefano
Recent works have proven that many relevant visual tasks are closely related one to another.
1 code implementation • CVPR 2019 • Alessio Tonioni, Oscar Rahnama, Thomas Joy, Luigi Di Stefano, Thalaiyasingam Ajanthan, Philip H. S. Torr
Real world applications of stereo depth estimation require models that are robust to dynamic variations in the environment.
no code implementations • 2 Feb 2019 • Alessio Tonioni, Luigi Di Stefano
Moreover, there exist a significant domain shift between the images that should be recognized at test time, taken in stores by cheap cameras, and those available for training, usually just one or a few studio-quality images per product.
1 code implementation • 29 Oct 2018 • Tommaso Cavallari, Stuart Golodetz, Nicholas A. Lord, Julien Valentin, Victor A. Prisacariu, Luigi Di Stefano, Philip H. S. Torr
The adapted forests achieved relocalisation performance that was on par with that of offline forests, and our approach was able to estimate the camera pose in close to real time.
no code implementations • 13 Oct 2018 • Pierluigi Zama Ramirez, Alessio Tonioni, Luigi Di Stefano
To prove the effectiveness of our proposal, we show how a semantic segmentation CNN trained on images from the synthetic GTA dataset adapted by our method can improve performance by more than 16% mIoU with respect to the same model trained on synthetic images.
1 code implementation • CVPR 2019 • Alessio Tonioni, Fabio Tosi, Matteo Poggi, Stefano Mattoccia, Luigi Di Stefano
Deep convolutional neural networks trained end-to-end are the state-of-the-art methods to regress dense disparity maps from stereo pairs.
2 code implementations • 10 Oct 2018 • Daniele De Gregorio, Gianluca Palli, Luigi Di Stefano
While robotic manipulation of rigid objects is quite straightforward, coping with deformable objects is an open issue.
1 code implementation • 9 Oct 2018 • Pierluigi Zama Ramirez, Matteo Poggi, Fabio Tosi, Stefano Mattoccia, Luigi Di Stefano
For unsupervised training of these models, geometry has been effectively exploited by suitable images warping losses computed from views acquired by a stereo rig or a moving camera.
no code implementations • 3 Oct 2018 • Alessio Tonioni, Eugenio Serra, Luigi Di Stefano
Then, available product databases usually include just one or a few studio-quality images per product (referred to herein as reference images), whilst at test time recognition is performed on pictures displaying a portion of a shelf containing several products and taken in the store by cheap cameras (referred to as query images).
1 code implementation • ICCV 2017 • Alessio Tonioni, Matteo Poggi, Stefano Mattoccia, Luigi Di Stefano
Recent ground-breaking works have shown that deep neural networks can be trained end-to-end to regress dense disparity maps directly from image pairs.
no code implementations • 26 Jul 2017 • Alessio Tonioni, Luigi Di Stefano
The arrangement of products in store shelves is carefully planned to maximize sales and keep customers happy.
1 code implementation • 19 Apr 2017 • Daniele De Gregorio, Luigi Di Stefano
We present a novel mapping framework for robot navigation which features a multi-level querying system capable to obtain rapidly representations as diverse as a 3D voxel grid, a 2. 5D height map and a 2D occupancy grid.
no code implementations • CVPR 2017 • Tommaso Cavallari, Stuart Golodetz, Nicholas A. Lord, Julien Valentin, Luigi Di Stefano, Philip H. S. Torr
Camera relocalisation is an important problem in computer vision, with applications in simultaneous localisation and mapping, virtual/augmented reality and navigation.
no code implementations • ICCV 2015 • Samuele Salti, Federico Tombari, Riccardo Spezialetti, Luigi Di Stefano
Keypoint detection represents the first stage in the majority of modern computer vision pipelines based on automatically established correspondences between local descriptors.
no code implementations • 13 Nov 2015 • Tommaso Cavallari, Luigi Di Stefano
Research works on the two topics of Semantic Segmentation and SLAM (Simultaneous Localization and Mapping) have been following separate tracks.
no code implementations • CVPR 2015 • Nicola Fioraio, Jonathan Taylor, Andrew Fitzgibbon, Luigi Di Stefano, Shahram Izadi
Our method supports online model correction, without needing to reprocess or store any input depth data.
no code implementations • CVPR 2013 • Samuele Salti, Alessandro Lanza, Luigi Di Stefano
The paper conjectures and demonstrates that repeatable keypoints based on salient symmetries at different scales can be detected by a novel analysis grounded on the wave equation rather than the heat equation underlying traditional Gaussian scale-space theory.
no code implementations • CVPR 2013 • Nicola Fioraio, Luigi Di Stefano
In this paper we propose a novel Semantic Bundle Adjustment framework whereby known rigid stationary objects are detected while tracking the camera and mapping the environment.