3DOR 2022 - Short Papers

Permanent URI for this collection

https://diglib7.eg.org/handle/10.2312/2633210

Browse

Browsing 3DOR 2022 - Short Papers by Issue Date

Filter results by year or month
Now showing 1 - 5 of 5
  • Thumbnail Image
    Item
    Labeled Facets: New Surface Texture Dataset
    (The Eurographics Association, 2022) Ganapathi, Iyyakutti Iyappan; Werghi, Naoufel; Berretti, Stefano; Thehoaris, Theoharis; Daoudi, Mohamed; Ferrari, Claudio; Veltkamp, Remco C.
    Object detection, recognition, segmentation, and retrieval have been at the forefront of 2D and 3D computer vision for a long time and have been utilized to address various problems in interdisciplinary domains. The 3D domain has not received as much attention as the 2D domain in several of these fields, and texture analysis in 3D is one of the least investigated. In the literature, there are several classic methods for retrieving and classifying 3D textures; however, research on facet-wise texture classification and segmentation is sparse. Moreover, in recent years deep learning excels in computer vision; utilizing its capacity for 3D texture analysis could improve performance compared to classical approaches. However, the scarcity of 3D texture data makes it challenging to employ deep learning. This paper presents a labeled 3D dataset based on already existing 3D datasets that can be utilized for texture classification, segmentation, and detection. The textures in the dataset are varied, with a wide range of surface variations. The dataset provides 3D texture surfaces annotated at the facet level, as well as fundamental geometric attributes such as curvature and shape index that can be utilized directly for further analysis. Download link for the dataset https://bit.ly/3wgSQgW.
  • Thumbnail Image
    Item
    Reconstructing 3D Face of Infants in Social Interactions Using Morphable Models of Non-Infants
    (The Eurographics Association, 2022) Sariyanidi, Evangelos; Zampella, Casey J.; Drye, Madison N.; Fecher, Madison L.; Magginson, Grace; Cubit, Laura Soskey; Schultz, Robert T.; Guthrie, Whitney; Tunc, Birkan; Berretti, Stefano; Thehoaris, Theoharis; Daoudi, Mohamed; Ferrari, Claudio; Veltkamp, Remco C.
    3D morphable models (3DMMs) simultaneously reconstruct facial morphology, expression and pose from 2D images, and thus could be an invaluable tool for capturing and characterizing the face and facial behavior in early childhood. However, 3DMM fitting on infants is a largely unexplored problem. All publicly available 3DMMs are developed for adults, and it is unclear if and to what extent they can be used on videos of infants. In this paper, we compare five state-of-the-art 3DMM fitting methods on data from naturalistic infant-caregiver interactions. Results suggest that it is possible to produce consistent and subject-specific reconstructions of 3D shape identity from multiple frames, but not from a single frame. Qualitative evaluation highlights that facial regions with high texture variation, such as eyes, brows and mouth, are captured with higher accuracy compared to the rest of the face. Thus, even though a 3DMM developed for adults has significant limitations when reconstructing the morphology of the entire facial region of infants, applications that involve analysis of facial behavior can be feasible. Our encouraging results, combined with the unique ability of 3DMMs to disentangle two major sources of noise for expression analysis (i.e., identity bias and pose variations), motivate future research on using 3DMMs to measure the facial behavior of infants.
  • Thumbnail Image
    Item
    Eurographics Workshop on 3D Object Retrieval - Short Papers: Frontmatter
    (The Eurographics Association, 2022) Berretti, Stefano; Thehoaris, Theoharis; Daoudi, Mohamed; Ferrari, Claudio; Veltkamp, Remco C.; Berretti, Stefano; Thehoaris, Theoharis; Daoudi, Mohamed; Ferrari, Claudio; Veltkamp, Remco C.
  • Thumbnail Image
    Item
    Parameterization Robustness of 3D Auto-Encoders
    (The Eurographics Association, 2022) Pierson, Emery; Besnier, Thomas; Daoudi, Mohamed; Arguillère, Sylvain; Berretti, Stefano; Thehoaris, Theoharis; Daoudi, Mohamed; Ferrari, Claudio; Veltkamp, Remco C.
    The generation of 3-dimensional geometric objects in the most efficient way is a thriving research topic with, for example, the development of geometric deep learning, extending classical machine learning concepts to non euclidean data such as graphs or meshes. In this short paper, we study the effect of a reparameterization on two popular mesh and point cloud neural networks in an auto-encoder mode: PointNet [QSMG16] and SpiralNet [BBP∗19]. Finally, we tested a modified version of PointNet that takes orientation into account (through coordinates of the normals) as a first step towards the construction of a geometric deep learning model built with a more flexible metric regarding the parameterization. The experimental results on standardized face datasets show that SpiralNet is more robust to the reparametrization than PointNet in this specific context with the proposed reparameterization.
  • Thumbnail Image
    Item
    Single Shot Phase Shift 3D Scanning with Convolutional Neural Network and Synthetic Fractals
    (The Eurographics Association, 2022) Li, Ke; Spehr, Marcel; Höhne, Daniel; Bräuer-Burchardt, Christian; Tünnermann, Andreas; Kühmstedt, Peter; Berretti, Stefano; Thehoaris, Theoharis; Daoudi, Mohamed; Ferrari, Claudio; Veltkamp, Remco C.
    The phase shift algorithm is an important 3D shape reconstruction method in industrial quality inspection and reverse engineering. To retrieve dense and accurate point clouds, the conventional phase shift methods require at least three fringe projection patterns, limiting its application to statics or semi-statics scenes only. In this paper, we propose a novel and low-cost single-shot phase shift 3D reconstruction framework using convolution neural networks (CNN) trained on 3D synthetic fractals. We first design and optimize a novel projection pattern that compresses the phase period orders and the ambiguous phase information into a single image. Then, we train two different CNNs to predict the ambiguous phase information and the period orders separately. The CNNs were trained on randomly generated 3D shapes whose geometric complexity is modeled by recursive shape generation algorithms which can create an unlimited amount of random 3D shapes on the fly. Initial results demonstrate that our method can produce high-quality point clouds from just a pair of 2D images, thus improving the temporal resolution of a phase-shift 3D scanner to the highest possible. As we also include different real-world lighting and textural conditions in the training data set, experiments also demonstrate that our CNN models which were trained on random synthetic fractals only can perform equally well in the real world.