Italian Chapter Conference
Permanent URI for this community
Browse
Browsing Italian Chapter Conference by Title
Now showing 1 - 20 of 327
Results Per Page
Sort Options
Item A 3 Cent Recognizer: Simple and Effective Retrieval and Classification of Mid-air Gestures from Single 3D Traces(The Eurographics Association, 2017) Caputo, Fabio Marco; Prebianca, Pietro; Carcangiu, Alessandro; Spano, Lucio D.; Giachetti, Andrea; Andrea Giachetti and Paolo Pingi and Filippo StancoIn this paper we present a simple 3D gesture recognizer based on trajectory matching, showing its good performances in classification and retrieval of command gestures based on single hand trajectories. We demonstrate that further simplifications in porting the classic "1 dollar" algorithm approach from the 2D to the 3D gesture recognition and retrieval problems can result in very high classification accuracy and retrieval scores even on datasets with a large number of different gestures executed by different users. Furthermore, recognition can be good even with heavily subsampled path traces and with incomplete gestures.Item 3-SHIRT: Three-Dimensional Shape Indexing and Retrieval Techniques(The Eurographics Association, 2008) Castellani, Umberto; Cortelazzo, Guido Maria; Cristani, M.; Delponte, Elisabetta; Fusiello, Andrea; Giachetti, Andrea; Mizzaro, Stefano; Odone, Francesca; Puppo, Enrico; Scateni, Riccardo; Zanuttigh, Pietro; Vittorio Scarano and Rosario De Chiara and Ugo ErraThis paper describes the work that has been done during the first year of the 3-SHIRT project, which aims at developing innovative solutions in all the phases of content-based 3D shape retrieval, namely: shape analysis and segmentation, design of shape descriptors, shape indexing and matching, and evaluation.Item 3D Body Model Construction and Matching for Real Time People Re-Identification(The Eurographics Association, 2010) Baltieri, Davide; Vezzani, Roberto; Cucchiara, Rita; Enrico Puppo and Andrea Brogni and Leila De FlorianiWide area video surveillance always requires to extract and integrate information coming from different cameras and views. Re-identification of people captured from different cameras or different views is one of most challenging problems. In this paper, we present a novel approach for people matching with vertices-based 3D human models. People are detected and tracked in each calibrated camera, and their silhouette, appearance, position and orientation are extracted and used to place, scale and orientate a 3D body model. Colour features are computed from the 2D appearance images and mapped to the 3D model vertices, generating the 3D model for each tracked person. A distance function between 3D models is defined in order to find matches among models belonging to the same person. This approach achieves robustness against partial occlusions, pose and viewpoint changes. A first experimental evaluation is conducted using images extracted from a real camera set-up.Item 3D City Reconstruction from OpenStreetMap Data(The Eurographics Association, 2021) Kaszuba, Sara; Pellacini, Fabio; Frosini, Patrizio and Giorgi, Daniela and Melzi, Simone and Rodolà , EmanueleVirtual city generation from real data is far from being straightforward for users, as it strictly depends on the application domain, amount of information available, and the adopted reconstruction techniques. Nowadays, reconstruction of virtual cities is of interests in entertainment, urban planning, emergency response and machine learning. To serve these applications, we have developed an open-source tool that can reconstruct cities at scale directly from OpenStreetMap data, that can perform full city generation in the order of hundreds of seconds.Item 3D CT Segmentation for Clinical Evaluation of Knee Prosthesis Operations(The Eurographics Association, 2006) Battiato, S.; Bosco, C.; Farinella, G. M.; Impoco, G.; S. Battiato and G. Gallo and F. StancoIn surgical knee replacement, the damaged knee joint is replaced with artificial prostheses. An accurate clinical evaluation must be carried out before applying knee prostheses to ensure optimal outcome from surgical operations and to reduce the probability of having long-term problems. Useful information can be inferred from CT scans of the knee area, that can be used to estimate the stress that will be acting on the prosthesis being implanted. This information can be exploited to tailor the prothesis to the patient s anatomy. We present some ideas for a system intended to help surgeons in evaluating the post-operative outcome of knee prosthesis implantation. We also show some preliminary results about the 3D segmentation of CT data, that is the main activity of our lab in the project.Item 3D data segmentation using a non-parametric density estimation approach(The Eurographics Association, 2006) Castellani, U.; Cristani, M.; Murino, V.; S. Battiato and G. Gallo and F. StancoIn this paper, a new segmentation approach for sets of 3D unorganized points is proposed. The method is based on a clustering procedure that separates the modes of a non-parametric multimodal density, following the mean-shift paradigm. The main idea consists in projecting the source 3D data into a set of independent sub-spaces, forming a joint multidimensional space. Each sub-space describes a geometric aspect of the data set, such as the normals and principal curvatures, so as a dense region in a particular sub-space indicates a set of 3D points sharing a similar value of that feature. A non-parametric clustering method is applied in this joint space by using a multidimensional kernel. This kernel smoothly takes into account for all the subspaces, moving towards high density regions in the joint space, separating them and providing "natural" clusters of 3D points. The algorithm can be implemented very easily and only few parameters need to be freely tuned. Experiments are reported, both on synthetic and real data, assessing the quality of the proposed approach and promoting further developments.Item 3D Digital Imaging for Knowledge Dissemination of Greek Archaic Statuary(The Eurographics Association, 2016) Stanco, Filippo; Tanasi, Davide; Allegra, Dario; Milotta, Filippo L. M.; Giovanni Pintore and Filippo StancoThis paper aims, using a research exercise, to verify the association between two Greek sculptures collected at different times: the head of a boy collected in the Chalcidian colony of Leontinoi in southeastern Sicily, acquired in the 18th century and later kept in the collection of the Museum of Castello Ursino in Catania, and a torso, retrieved in 1904 and since then displayed in the Archaeological Museum of Sicily. The two pieces share similar stylistic features and represent the most significant example of Greek sculpture in Sicily at the end of the 6th century BC. Their association is an open problem still debated by scholars, who have based their studies on comparisons between pictures as a reassembly of two artefacts was never attempted. This critical issue has conditioned curators of the two museums, who could not develop a proper communication policy for the two objects, resulting in a limited cognitive accessibility for the public. By means of 3D scanning techniques, this contribution showcases how virtual restoration can not only improve interpretations of the scholars, but also boost the communication plans of museums, giving back to the public via a web platform a masterpiece of Greek sculpture known just by specialists.Item 3D Input devices integration in CAD environment(The Eurographics Association, 2006) Bruno, F.; Caruso, G.; Muzzupappa, M.; S. Battiato and G. Gallo and F. StancoVirtual Reality (VR) technologies are becoming commonly used tools in the product development process, starting from the styling review in the conceptual design phase, until to the Digital Mock-up (DMU) validation in the advanced stages of the design process. What has not yet been sufficiently investigated is the possibility to interact with the DMU directly inside the CAD environment using 3D input devices. Although few CAD systems, like CATIA V5, have an additional module to support VR devices, in most cases it is still necessary to customize the application to obtain the functionalities desired by the user. The present paper discusses difficulties and advantages related to the integration of VR techniques and 3D input devices in CAD systems. The work has been conducted analyzing and testing the potentialities of the Unigraphics NX3 CAD environment and implementing a software plug-in that allows the user to perform such interaction tasks employing 3D input devices.Item 3D Modeling and Integration of Heterogeneous Geo-data(The Eurographics Association, 2021) Miola, Marianna; Cabiddu, Daniela; Pittaluga, Simone; Mortara, Michela; Vetuschi Zuccolini, Marino; Imitazione, Gianmario; Frosini, Patrizio and Giorgi, Daniela and Melzi, Simone and Rodolà , EmanueleThis paper tackles the volumetric representation of geophysical and geotechnical data, gathered during exploration surveys of the subsoil. The creation of a 3D model as support to geological interpretation has to take into account the specificity of the diverse input data, that are heterogeneous. Some data are massive, but cover the domain unevenly, e.g., structured along dense differently spaced lines, while others are very sparse, e.g., borehole locations with soil sampling and CPTU (Piezocone Penetration Test) locations. In this work, we focus on the exploration and analysis of underwater deposits. After a discussion about the data typically acquired in an offshore campaign, we present an automatic process to generate the subsurfaces and volume defining an underground deposit, starting from the identification of relevant morphological features in seismic data. In particular, data simplification and refinement based on geostatistics have been applied to generate regular 2D meshes from strongly anisotropic data, in order to improve the quality of the final 3D tetrahedral mesh. Furthermore, we also use geostatistics to predict geotechnical parameters from local surveys and estimate their distribution on the whole domain: in this way the 3D model will include relevant geological features of the deposit and allow extrapolating different geotechnical information with associated uncertainty. The volume characterization and its 3D inspection will improve the structural and stratigraphic interpretation of deposits, to support geological analysis and planning of future engineering activities.Item 3D Objects Face Clustering using Unsupervised Mean Shift(The Eurographics Association, 2007) Farenzena, M.; Cristani, M.; Castellani, U.; Fusiello, A.; Raffaele De Amicis and Giuseppe ContiIn this paper, a novel approach to face clustering is proposed. The aim is the completely unsupervised extraction of planes in a polygonal a mesh, obtained from a 3D reconstruction process. In this context, 3D coordinates points are inevitably affected by error, therefore resiliency is a primal concern in the analysis. The method is based on the Mean Shift clustering paradigm, devoted to separating modes of a multimodal non-parametric density, by using a kernel-based technique. A critical parameter, the kernel bandwidth size, is here automatically detected by following a well-accepted partition stability criterion. Experimental and comparative results on synthetic and real data validate the approach.Item 3D Reconstruction and Augmented Reality in Bronchoscopic Intervention(The Eurographics Association, 2011) Torrisi, Alessandro; Livatino, Salvatore; Gallo, Giovanni; Andrea F. Abate and Michele Nappi and Genny TortoraThe use of stereoscopic visualization has recently been proposed for many applications. Other than for entertainment, stereo viewing is being proposed for robotics and medical teleoperation. This paper proposes to exploit the stereo-camera setup available in stereo-viewing systems for 3D reconstruction. We focus on medical endoscopic applications. The advantages of having a reconstructed vision-based depth map during endoscopic navigation and intervention are many, including the possibility for generation of an augmented-reality visual scenario to support surgeons during interventions. We ran our experiments on a realistic graphic model of human bronchus to study feasibility of the proposed concept. We reconstructed depth maps of a bronchus environment and used them to generate augmented reality views of the observed scenes.Item 3D Reconstruction of Facade and the Statues of the Church ''Maria Santissima dell'Elemosina'' in Biancavilla(The Eurographics Association, 2017) Tomasello, Simone Federico; Pazzi, Miriam Caterina; Andrea Giachetti and Paolo Pingi and Filippo Stanco3D modelling plays a key role in the reconstruction of cultural heritage assets. It allows to give an idea of the original condition of a cultural heritage before time and weathering effects. This paper presents the 3D reconstruction of the church ''Maria Santissima dell'Elemosina'' through various software for modelling, as well as the acquisition and processing of elements. High resolution photos has been employed for faithful reproduction of the models details.Item 3D Scene Comparison using Topological Graphs(The Eurographics Association, 2007) Paraboschi, L.; Biasotti, S.; Falcidieno, B.; Raffaele De Amicis and Giuseppe ContiNew technologies for shape acquisition and rendering of digital shapes have simplified the process of creating virtual scenes; nonetheless, shape annotation, recognition and manipulation of both the complete virtual scenes and even of subparts of them are still open problems. In this paper we deal with the problem of comparing two (or more) object sets, where each model is represented by an attributed graph. We will define a new distance to estimate the possible similarities among the sets of graphs and will validate our work using the shape graph [BGSF06].Item 3D-ize U! A Real-time 3D Head-model Texture Generator for Android(The Eurographics Association, 2011) Boi, Stefano; Sorrentino, Fabio; Marras, Stefano; Scateni, Riccardo; Andrea F. Abate and Michele Nappi and Genny TortoraRecently, the number of applications developed for smartphones has dramatically increased; however, at the moment, applications having the purpose of creating and displaying 3D models are quite rare. The goal of this work is to build an application that allows the user to see the virtual three-dimensional representations of their friends and interact with them. The main challenge is to achieve results similar to those that a computer would produce, optimizing the process to deal with the constraints due to the technology used. Since there are no similar mobile applications, this work will make possible to create a base onto which realizing applications that have customized 3D models as a common feature.Item 3DReg-i-Net: Improving Deep Learning Based 3D Registration for a Robust Real-time Alignment of Small-scale Scans(The Eurographics Association, 2019) Lombardi, Marco; Riccardi, Andrea; Savardi, Mattia; Signoroni, Alberto; Agus, Marco and Corsini, Massimiliano and Pintus, RuggeroWe present 3DReg-i-Net, an improved deep learning solution for pairwise registration of 3D scans, which evolves the recently proposed 3DRegNet technique by Pais et al. This is one of the very first learning based algorithm aiming at producing the co-registration of two 3D views starting solely from a set of point correspondences, which is able to perform outlier rejection and to recover the registration matrix. We evolve the original method to face the challenging scenario of quick 3D modelling at small scales through the alignment of dense 3D views acquired at video frame-rate with a handheld scanner. We improve the system tracking robustness and alignment performance with a generalized input data augmentation. Moreover, working on suboptimal aspects of the original solution, we propose different improvements that lead to a redefinition of the training loss function. When tested on the considered scenario, the proposed 3DReg-i-Net significantly outperforms the prior solution in terms of accuracy of the estimated aligning transforms.Item Accurate Molecular Atom Selection in VR(The Eurographics Association, 2022) Molina, Elena; Vázquez, Pere-Pau; Cabiddu, Daniela; Schneider, Teseo; Allegra, Dario; Catalano, Chiara Eva; Cherchi, Gianmarco; Scateni, RiccardoTarget acquisition is a basic task that is part of almost any high-level interaction in 3D environments. Therefore, providing accurate selection is a necessity for most applications and games. When targets are small and scenes are cluttered, selection becomes inaccurate. This may lead to selecting the wrong elements which, apart from the time consumed, may become a frustrating experience. Besides the unintentional tremor, the button/trigger press for effectively selecting an element further reduces our stability, increasing the probability of an incorrect target acquisition. In this paper, we focus on molecular visualization and address the problem of selecting atoms, which are rendered as small spheres. We build upon previous progressive selection algorithms and present two alternatives that accelerate the selection of neighbors after an initial selection. We have implemented and analyzed such techniques through a formal user study and found that they were highly appreciated by the users. These selection methods may be suitable for other crowded environments beyond molecular visualization.Item Adaptive Clipping of Splats to Models with Sharp Features(The Eurographics Association, 2014) Ivo, Rafael; Ganovelli, Fabio; Vidal, Creto; Scopigno, Roberto; Andrea GiachettiSplat-based models are a good representation because of its absense of topology, making complex modeling operations easier, but keeping the same approximation ratio from triangular meshes. However corners cannot be properly represented by splats without clipping them. We present a new method for clipping splats in models with sharp features. Each splat is an ellipse equipped with a few parameters that allow to define how the ellipse can be clipped against a bidimensional rational Bézier curve and thus it can be used for all those surfaces that show a large number of edge features and different sampling rate around them. The simple and uniform data used to define the clipping curve makes easy the implementation in GPU. We designed and implemented an automatic computation of the clipping curves and a pipeline for sampling a generic surface with splats and render it. In this paper we show how this technique outperforms the previous clipping techniques in precision for objects such as mechanical parts and CAD- like models keeping the rendering speed.Item Adaptive Environmental Sampling: The Interplay Between Geostatistics and Geometry(The Eurographics Association, 2018) Berretta, S.; Cabiddu, D.; Pittaluga, S.; Mortara, M.; Spagnuolo, M.; Zuccolini, M. Vetuschi; Livesu, Marco and Pintore, Gianni and Signoroni, AlbertoIn environmental surveys a large sampling effort is required to produce accurate geostatistical maps representing the distribution of environmental variables, and the analysis of each sample is often expensive. Typically, the sample locations are completely specified in the survey design phase, prior to data-collection. Usually, the sampling points are located on a regular grid, or along directions that are selected with respect to any a-priori knowledge of the expert. No feedback is available during the survey. In this paper, we present a different sampling strategy, namely adaptive sampling. Our approach exploits geostatistics constructs in order to determine on-the fly the next best sample location. After initializing the system with few sampling points, an iterative routine predicts the variable distribution from the data sampled so far, and suggests the next sample to be acquired in order to optimize the uncertainty of the estimates. At every iteration a new sample is acquired, and the variable distribution map is refined, along with the uncertainty map related to that distribution. Our method allows to build a representation of the survey area as precise as the one provided by the traditional methods, but with less samples, thus reducing both time and costs of the survey. We show a preliminary evaluation of the adaptive strategy in the bi-dimensional case based on a synthetic scenario, and describe the generalization of these encouraging results to the full 3D domain in the concrete setting of water quality monitoring. A proper geometric representation of the three dimensional survey area, coupled with a proper visualization of distribution and related uncertainty, will provide real-time feedback during the survey.Item Adaptive Frame Rate Up-conversion with Motion Extraction from 3D Space for 3D Graphic Pipelines(The Eurographics Association, 2006) Falchetto, M.; Barone, M.; Pau, D.; S. Battiato and G. Gallo and F. StancoA novel 3D Graphic Pipeline for mobile handheld devices is introduced. It s able to achieve increased frame-rate at its output, high picture quality, improved temporal anti-aliasing, thanks to frame-rate upconversion algorithm based on primitive s motion extraction and efficient compensation. Classic rendering process doesn t exploit spatial and temporal coherence of the motion: each frame is rendered by using a lot of the pipeline resources inside geometry, rasteriser and fragment processing stages. Moreover each frame is computed independently from the temporal adjacent one with a brute force approach. Video based algorithms instead (e.g. MPEG type of processing), heavily exploit temporal, spatial and statistical coherency of the content being coded, relying on motion estimation to extract motion vectors from pixel domain during coding process and apply them during motion compensation decoding process. Previous works on 3D graphics were based on the use of impostors, sprite or coherent layers. Those approaches are too much complex and requires high computational power. Other ones based on frame averaging are too much simple: they achieve poor spatial and temporal quality and artifacts become clearly visible as the motion dynamic increases. Proposed approach instead exploits motion temporal coherency by extracting motion vectors from motions of visible primitives in screen space. Temporal coherence has been exploited between key frames using an adaptive motion compensated stage. This method is able to reconstruct the frames without introducing noticeable artifacts on final pictures. Results achieved are 10-15 dB better quality than simple temporal frame average and show a natural motion with no annoying artifacts like real video content shows.Item Adaptive Quad Mesh Simplification(The Eurographics Association, 2010) Bozzo, Agostino; Panozzo, Daniele; Puppo, Enrico; Pietroni, Nico; Rocca, Luigi; Enrico Puppo and Andrea Brogni and Leila De FlorianiWe present an improved algorithm for the progressive simplification of quad meshes, which adapts the resolution of the mesh to details of the modeled shape. We extend previous work [TPC*10], by simplifying the approach and combining it with the concept of Fitmaps introduced in [PPT*10]. The new algorithm has several advantages: it is simpler and more robust; it does not need a parametrization of the input shape; it is adaptive; and it preserves projectability of the output mesh to the input shape, thus supporting displacement mapping. We present experimental results on a variety of datasets, showing relevant improvement over previous results under several aspects.