Search

Point cloud registration is a fundamental task in 3D reconstruction and environment perception. We explore the performance of modern Deep Learning-based registration techniques, in particular Deep Global Registration (DGR) and Learning Multiview Registration (LMVR), on an outdoor real world data consisting of thousands of range maps of a building acquired by a Velodyne LIDAR mounted on a drone. We used these pairwise registration methods in a sequential pipeline to obtain an initial rough registration. The output of this pipeline can be further globally refined. This simple registration pipeline allow us to assess if these modern methods are able to deal with this low quality data. Our experiments demonstrated that, despite some design choices adopted to take into account the peculiarities of the data, more work is required to improve the results of the registration.

We present the ongoing effort to build the first benchmark dataset for aestethic prediction on 3D models. The dataset is built on top of Sketchfab, a popular platform for 3D content sharing. In our dataset, the visual 3D content is aligned with aestheticsrelated metadata: each 3D model is associated with a number of snapshots taken from different camera positions, the number of times the model has been viewed in-between its upload and its retrieval, the number of likes the model got, and the tags and comments received from users. The metadata provide precious supervisory information for data-driven research on 3D visual attractiveness and preference prediction. The paper contribution is twofold. First, we introduce an interactive platform for visualizing data about Sketchfab. We report a detailed qualitative and quantitative analysis of numerical scores (views and likes collected by 3D models) and textual information (tags and comments) for different 3D object categories. The analysis of the content of Sketchfab provided us the base for selecting a reasoned subset of annotated models. The second contribution is the first version of the ViDA 3D dataset, which contains the full set of content required for data-driven approaches to 3D aesthetic analysis. While similar datasets are available for images, to our knowledge this is the first attempt to create a benchmark for aestethic prediction for 3D models. We believe our dataset can be a great resource to boost research on this hot and far-from-solved problem., Smart Tools and Apps for Graphics - Eurographics Italian Chapter Conference, Tools, 45, 55, Mattia Angelini, Vito Ferrulli, Francesco Banterle, Massimiliano Corsini, Maria Antonietta Pascali, Paolo Cignoni, and Daniela Giorgi, CCS Concepts: Computing methodologies --> Shape analysis

The automatic recognition of natural structures is a challenging task in the supervised learning field. Complex morphologies are difficult to detect both from the networks, that may suffer from generalization issues, and from human operators, affecting the consistency of training datasets. The task of manual annotating biological structures is not comparable to a generic task of detecting an object (a car, a cat, or a flower) within an image. Biological structures are more similar to textures, and specimen borders exhibit intricate shapes. In this specific context, manual labelling is very sensitive to human error. The interactive validation of the predictions is a valuable resource to improve the network performance and address the inaccuracy caused by the lack of annotation consistency of human operators reported in literature. The proposed tool, inspired by the Yes/No Answer paradigm, integrates the semantic segmentation results coming from a CNN with the previous human labeling, allowing a more accurate annotation of thousands of instances in a short time. At the end of the validation, it is possible to obtain corrected statistics or export the integrated dataset and re-train the network., Eurographics 2019 - Short Papers, Learning and Networks, 57, 60, Gaia Pavoni, Massimiliano Corsini, Marco Palma, and Roberto Scopigno, CCS Concepts: Human-centered computing --> Graphical user interfaces; Visual analytics; Computing methodologies --> Image segmentation

This document provides an in-depth analysis of visualization and interaction tools employed in the context of Virtual Museum. This analysis is required to identify and design the tools and the different components that will be part of the Common Implementation Framework (CIF). The CIF will be the base of the web-based services and tools to support the development of Virtual Museums with particular attention to online Virtual Museum. The main goal is to provide to the stakeholders and developers an useful platform to support and help them in the development of their projects, despite the nature of the project itself. The design of the Common Implementation Framework (CIF) is based on an analysis of the typical workflow of the V-MUST partners and their perceived limitations of current technologies. This document is based also on the results of the V-MUST technical questionnaire (presented in Deliverable 4.1). Based on these two sources of information, we have selected some important tools (mainly visualization tools) and services and we elaborate some first guidelines and ideas for the design and development of the CIF, that shall provide a technological foundation for the V-MUST Platform, together with the V-MUST repository/repositories and the additional services defined in the WP4. Two state-of-the-art reports, one about user interface design and another one about visualization technologies have been also provided in this document.

The dichotomy between full detail representation and the efficient management of data digitization is still a big issue in the context of the acquisition and visualization of 3D objects, especially in the field of the cultural heritage. Modern scanning devices enable very detailed geometry to be acquired, but it is usually quite hard to apply these technologies to large artifacts. In this article we present a project aimed at virtually reconstructing the impressive (7 × 11 m.) portal of the Ripoll Monastery, Spain. The monument was acquired using triangulation laser scanning technology, producing a dataset of 2212 range maps for a total of more than 1 billion triangles. All the steps of the entire project are described, from the acquisition planning to the final setup for dissemination to the public. We show how time-of-flight laser scanning data can be used to speed-up the alignment process. In addition we show how, after creating a model and repairing imperfections, an interactive and immersive setup enables the public to navigate and display a fully detailed representation of the portal. This article shows that, after careful planning and with the aid of state-of-the-art algorithms, it is now possible to preserve and visualize highly detailed information, even for very large surfaces.

International audience; Recent advances in 3D graphics technologies have led to an increasing use of processing techniques on 3D meshes,such as filtering, compression, watermarking, simplification, deformation and so forth. Since these processes may modify thevisual appearance of the 3D objects, several metrics have been introduced so as to properly drive or evaluate them, from classicgeometric ones such as Hausdorff distance, to more complex perceptually-based measures. This paper presents a survey on existingperceptually-based metrics for visual impairment of 3D objects and provides an extensive comparison between them. In particular,different scenarios which correspond to different perceptual and cognitive mechanisms are analyzed. The objective is twofold: (1)catching the behavior of existing measures so as to help Perception researchers for designing new 3D metrics and (2) providing acomparison between them so as to inform and help Computer Graphics researchers for choosing the accurate tool for the design andthe evaluation of their mesh processing algorithms.

3D scanning technologies offer a lot of interesting possibilities for uses related to Cultural Heritage. Unfortunately, most of the current 3D scanning solutions are very costly and require a significant investment, both in terms of software and hardware. In this paper we discuss how to perform 3D scanning for the acquisition of Cultural Heritage objects using only low cost hardware and open source or free software tools.