Search Results

Now showing 1 - 10 of 16
  • Item
    An Interactive Tuning Method for Generator Networks Trained by GAN
    (The Eurographics Association, 2022) Zhou, Mengyuan; Yamaguchi, Yasushi; Cabiddu, Daniela; Schneider, Teseo; Allegra, Dario; Catalano, Chiara Eva; Cherchi, Gianmarco; Scateni, Riccardo
    The recent studies on GAN achieved impressive results in image synthesis. However, they are still not so perfect that output images may contain unnatural regions. We propose a tuning method for generator networks trained by GAN to improve their results by interactively removing unexpected objects and textures or changing the object colors. Our method could find and ablate those units in the generator networks that are highly related to the specific regions or their colors. Compared to the related studies, our proposed method can tune pre-trained generator networks without relying on any additional information like segmentation-based networks. We built the interactive system based on our method, capable of tuning the generator networks to make the resulting images as expected. The experiments show that our method could remove only unexpected objects and textures. It could change the selected area color as well. The method also gives us some hints to discuss the properties of generator networks which layers and units are associated with objects, textures, or colors.
  • Item
    Creating Adaptive and Interactive Stories in Mixed Reality
    (The Eurographics Association, 2022) Frau, Vittoria; Serra, Sergio; Spano, Lucio Davide; Cabiddu, Daniela; Schneider, Teseo; Allegra, Dario; Catalano, Chiara Eva; Cherchi, Gianmarco; Scateni, Riccardo
    The following paper proposes the study, the design and the preliminary development of a solution for supporting users without programming experience in creating stories in a Mixed Reality environment. We focus on a Mixed Reality interface split into two parts: the creation and observation phases. During the creation phase, the end user can build his/her own story in the immersive mode of the Mixed Reality experience. The user can also enjoy the stories that other users have designed by seeing the characters appear in their surrounding environment.
  • Item
    IMGD: Image-based Multiscale Global Descriptors of Airborne LiDAR Point Clouds Used for Comparative Analysis
    (The Eurographics Association, 2021) Sreevalsan-Nair, Jaya; Mohapatra, Pragyan; Singh, Satendra; Frosini, Patrizio and Giorgi, Daniela and Melzi, Simone and Rodolà, Emanuele
    Both geometric and semantic information are required for a complete understanding of regions acquired as three-dimensional (3D) point clouds using the Light Detection and Ranging (LiDAR) technology. However, the global descriptors of such datasets that integrate both the information types are rare. With a focus on airborne LiDAR point clouds, we propose a novel global descriptor that transforms the point cloud from Cartesian to barycentric coordinate spaces. We use both the probabilistic geometric classification, aggregated from multiple scales, and the semantic classification to construct our descriptor using point rendering. Thus, we get an image-based multiscale global descriptor, IMGD. To demonstrate its usability, we propose the use of distribution distance measures between the descriptors for comparing the point clouds. Our experimental results demonstrate the effectiveness of our descriptor, when constructed of publicly available datasets, and on applying our selected distance measures.
  • Item
    ReVize: A Library for Visualization Toolchaining with Vega-Lite
    (The Eurographics Association, 2019) Hogräfer, Marius; Schulz, Hans-Jörg; Agus, Marco and Corsini, Massimiliano and Pintus, Ruggero
    The field of tools for data visualization has been growing in recent years, with each tool contributing new ways to create and work with visualizations, and each offering a specialized set of features, interaction metaphors and user interfaces. This means on one hand that users have a wide choice in visualization tools. On the other hand, though, this choice might also lock-in the user: Once made, it becomes difficult and sometimes even impossible to switch to another tool - e.g., to further refine a visualization made in one tool inside another. In turn, users are forced to work around any shortcomings of the chosen tool, as switching to another tool is even more cumbersome. In this paper, we introduce ReVize, an open-source library for visualization toolchaining. ReVize makes use of Vega-Lite as a common exchange format to be able to add toolchain support to web-based tools. In contrast to existing approaches, this solution to visualization toolchaining allows for authoring a visualization with multiple tools in a back-and-forth fashion, without a preset order in which tools are to be used. We demonstrate ReVize by adding toolchain support to three existing tools - KNIME, ColorBrewer, and VisFlow - for using them in concert to author visualizations.
  • Item
    Motion Data and Model Management for Applied Statistical Motion Synthesis
    (The Eurographics Association, 2019) Herrmann, Erik; Du, Han; Antakli, André; Rubinstein, Dmitri; Schubotz, René; Sprenger, Janis; Hosseini, Somayeh; Cheema, Noshaba; Zinnikus, Ingo; Manns, Martin; Fischer, Klaus; Slusallek, Philipp; Agus, Marco and Corsini, Massimiliano and Pintus, Ruggero
    Machine learning based motion modelling methods such as statistical modelling require a large amount of input data. In practice, the management of the data can become a problem in itself for artists who want to control the quality of the motion models. As a solution to this problem, we present a motion data and model management system and integrate it with a statistical motion modelling pipeline. The system is based on a data storage server with a REST interface that enables the efficient storage of different versions of motion data and models. The database system is combined with a motion preprocessing tool that provides functions for batch editing, retargeting and annotation of the data. For the application of the motion models in a game engine, the framework provides a stateful motion synthesis server that can load the models directly from the data storage server. Additionally, the framework makes use of a Kubernetes compute cluster to execute time consuming processes such as the preprocessing and modelling of the data. The system is evaluated in a use case for the simulation of manual assembly workers.
  • Item
    STRONGER: Simple TRajectory-based ONline GEsture Recognizer
    (The Eurographics Association, 2021) Emporio, Marco; Caputo, Ariel; Giachetti, Andrea; Frosini, Patrizio and Giorgi, Daniela and Melzi, Simone and Rodolà, Emanuele
    In this paper, we present STRONGER, a client-server solution for the online gesture recognition from captured hands' joints sequences. The system leverages a CNN-based recognizer improving current state-of-the-art solutions for segmented gestures classification, trained and tested for the online gesture recognition task on a recent benchmark including heterogeneous gestures. The recognizer provides good classification accuracy and a limited number of false positives on most of the gesture classes of the benchmark used and has been used to create a demo application in a Mixed Reality scenario using an Hololens 2 optical see through Head-Mounted Display with hand tracking capability.
  • Item
    Design and Implementation of a Visualization Tool for the in-depth Analysis of the Domestic Electricity Consumption
    (The Eurographics Association, 2019) Merlin, Gabriele; Ortu, Daniele; Cherchi, Gianmarco; Scateni, Riccardo; Agus, Marco and Corsini, Massimiliano and Pintus, Ruggero
    In this poster, we present a visualization tool for the in-depth analysis of domestic electricity consumption. The web-interface allows users to visualize their electricity consumption, compare them with their own records or with the means of selected communities.
  • 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, Riccardo
    Target 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
    Visual Representation of Region Transitions in Multi-dimensional Parameter Spaces
    (The Eurographics Association, 2019) Fernandes, Oliver; Frey, Steffen; Reina, Guido; Ertl, Thomas; Agus, Marco and Corsini, Massimiliano and Pintus, Ruggero
    We propose a novel visual representation of transitions between homogeneous regions in multi-dimensional parameter space. While our approach is generally applicable for the analysis of arbitrary continuous parameter spaces, we particularly focus on scientific applications, like physical variables in simulation ensembles. To generate our representation, we use unsupervised learning to cluster the ensemble members according to their mutual similarity. In doing this, clusters are sorted such that similar clusters are located next to each other. We then further partition the clusters into connected regions with respect to their location in parameter space. In the visualization, the resulting regions are represented as glyphs in a matrix, indicating parameter changes which induce a transition to another region. To unambiguously associate a change of data characteristics to a single parameter, we specifically isolate changes by dimension. With this, our representation provides an intuitive visualization of the parameter transitions that influence the outcome of the underlying simulation or measurement. We demonstrate the generality and utility of our approach on diverse types of data, namely simulations from the field of computational fluid dynamics and thermodynamics, as well as an ensemble of raycasting performance data.
  • Item
    An Information-theoretic Visual Analysis Framework for Convolutional Neural Networks
    (The Eurographics Association, 2021) Shen, Jingyi; Shen, Han-Wei; Frosini, Patrizio and Giorgi, Daniela and Melzi, Simone and Rodolà, Emanuele
    Despite the great success of Convolutional Neural Networks (CNNs) in Computer Vision and Natural Language Processing, the working mechanism behind CNNs is still under extensive discussion and research. Driven by strong demand for the theoretical explanation of neural networks, some researchers utilize information theory to provide insight into the black-box model. However, to the best of our knowledge, employing information theory to quantitatively analyze and qualitatively visualize neural networks has not been extensively studied in the visualization community. In this paper, we combine information entropies and visualization techniques to shed light on how CNN works. Specifically, we first introduce a data model to organize the data that can be extracted from CNN models. Then we propose two ways to calculate entropy under different circumstances. To provide a fundamental understanding of the basic building blocks of CNNs (e.g., convolutional layers, pooling layers, normalization layers) from an information-theoretic perspective, we develop a visual analysis system, CNNSlicer. CNNSlicer allows users to interactively explore the amount of information changes inside the model. With case studies on the widely used benchmark datasets (MNIST and CIFAR-10), we demonstrate the effectiveness of our system in opening the black-box of CNNs.