Browsing by Author "Biasotti, Silvia"

Now showing 1 - 6 of 6
  • Thumbnail Image
    Item
    Automatic Segmentation of Archaeological Fragments with Relief Patterns using Convolutional Neural Networks
    (The Eurographics Association, 2021) Thompson, Elia Moscoso; Ranieri, Andrea; Biasotti, Silvia; Hulusic, Vedad and Chalmers, Alan
    The recent commodification of high-quality 3D scanners is leading to the possibility of capturing models of archaeological finds and automatically recognizing their surface reliefs. We present our advancements in this field using Convolutional Neural Networks (CNNs) to segment and classify the region around a vertex in a robust way. The network is trained with high-resolution views of the 3D models captured at different angles. The views represent both the model with its original textures and a colorization of the patches according to the value of the Shape Index (SI) in their vertices. The SI encodes local surface variations and we exploit the colorization of the model driven by the SI to generate other view and enrich the dataset. Our method has been validated on a relief recognition benchmark on archaeological fragments proposed within the SHape REtrieval Contest (SHREC) 2018.
  • No Thumbnail Available
    Item
    Feature-based Characterisation of Patient-specific 3D Anatomical Models
    (The Eurographics Association, 2019) Banerjee, Imon; Paccini, Martina; Ferrari, Enrico; CATALANO, CHIARA EVA; Biasotti, Silvia; Spagnuolo, Michela; Agus, Marco and Corsini, Massimiliano and Pintus, Ruggero
    This paper aims to examine the potential of 3D shape analysis integrated to machine learning techniques in supporting medical investigation. In particular, we introduce an approach specially designed for the characterisation of anatomical landmarks on patient-specific 3D carpal bone models represented as triangular meshes. Furthermore, to identify functional articulation regions, two novel district-based properties are defined. The performance of both state of the art and novel features has been evaluated in a machine learning setting to identify a set of significant anatomical landmarks on patient data. Experiments have been performed on a carpal dataset of 56 patient-specific 3D models that are segmented from T1 weighed magnetic resonance (MR) scans of healthy male subjects. Despite the typical large inter-patient shape variation within the training samples, our framework has achieved promising results.
  • No Thumbnail Available
    Item
    A Geometric Approach for Computing the Kernel of a Polyhedron
    (The Eurographics Association, 2021) Sorgente, Tommaso; Biasotti, Silvia; Spagnuolo, Michela; Frosini, Patrizio and Giorgi, Daniela and Melzi, Simone and RodolĆ , Emanuele
    We present a geometric algorithm to compute the geometric kernel of a generic polyhedron. The geometric kernel (or simply kernel) is defined as the set of points from which the whole polyhedron is visible. Whilst the computation of the kernel for a polygon has already been largely addressed in the literature, less has been done for polyhedra. Currently, the principal implementation of the kernel estimation is based on the solution of a linear programming problem. We compare against it on several examples, showing that our method is more efficient in analysing the elements of a generic tessellation. Details on the technical implementation and discussions on pros and cons of the method are also provided.
  • No Thumbnail Available
    Item
    HT-based Recognition of Patterns on 3D Shapes Using a Dictionary of Mathematical Curves
    (The Eurographics Association, 2019) Romanengo, Chiara; Biasotti, Silvia; FALCIDIENO, BIANCA; Agus, Marco and Corsini, Massimiliano and Pintus, Ruggero
    Characteristic curves play a fundamental role in the way a shape is perceived and illustrated. To address the curve recognition problem on surfaces, we adopt a generalisation of the Hough Transform (HT) which is able to deal with mathematical curves. In particular, we extend the set of curves so far adopted for curve recognition with the HT and propose a new dictionary of curves to be selected as templates. In addition, we introduce rules of composition and aggregation of curves into patterns, not limiting the recognition to a single curve at a time. Our method recognises various curves and patterns, possibly compound on a 3D surface. It selects the most suitable profile in a family of curves and, deriving from the HT, it is robust to noise and able to deal with data incompleteness. The system we have implemented is open and allows new additions of curves in the dictionary of functions already available.
  • Thumbnail Image
    Item
    mpLBP: An Extension of the Local Binary Pattern to Surfaces based on an Efficient Coding of the Point Neighbours
    (The Eurographics Association, 2019) Moscoso Thompson, Elia; Biasotti, Silvia; Digne, Julie; Chaine, Raphaƫlle; Biasotti, Silvia and LavouƩ, Guillaume and Veltkamp, Remco
    The description of surface textures in terms of repeated colorimetric and geometric local surface variations is a crucial task for several applications, such as object interpretation or style identification. Recently, methods based on extensions to the surface meshes of the Local Binary Pattern (LBP) or the Scale-Invariant Feature Transform (SIFT) descriptors have been proposed for geometric and colorimetric pattern retrieval and classification. With respect to the previous works, we consider a novel LBPbased descriptor based on the assignment of the point neighbours into sectors of equal area and a non-uniform, multiple ring sampling. Our method is able to deal with surfaces represented as point clouds. Experiments on different benchmarks confirm the competitiveness of the method within the existing literature, in terms of accuracy and computational complexity.
  • No Thumbnail Available
    Item
    A Survey of Indicators for Mesh Quality Assessment
    (The Eurographics Association and John Wiley & Sons Ltd., 2023) Sorgente, Tommaso; Biasotti, Silvia; Manzini, Gianmarco; Spagnuolo, Michela; Bousseau, Adrien; Theobalt, Christian
    We analyze the joint efforts made by the geometry processing and the numerical analysis communities in the last decades to define and measure the concept of ''mesh quality''. Researchers have been striving to determine how, and how much, the accuracy of a numerical simulation or a scientific computation (e.g., rendering, printing, modeling operations) depends on the particular mesh adopted to model the problem, and which geometrical features of the mesh most influence the result. The goal was to produce a mesh with good geometrical properties and the lowest possible number of elements, able to produce results in a target range of accuracy. We overview the most common quality indicators, measures, or metrics that are currently used to evaluate the goodness of a discretization and drive mesh generation or mesh coarsening/refinement processes. We analyze a number of local and global indicators, defined over two- and three-dimensional meshes with any type of elements, distinguishing between simplicial, quadrangular/hexahedral, and generic polytopal elements. We also discuss mesh optimization algorithms based on the above indicators and report common libraries for mesh analysis and quality-driven mesh optimization.