3DOR 2023 - Short Papers

Permanent URI for this collection

https://diglib.eg.org/handle/10.2312/3543874

Browse

Browsing 3DOR 2023 - Short Papers by Title

Now showing 1 - 4 of 4
  • No Thumbnail Available
    Item
    Eurographics Workshop on 3D Object Retrieval - Short Papers: Frontmatter
    (The Eurographics Association, 2023) Fugacci, Ulderico; Lavoué, Guillaume; Veltkamp, Remco C.; Fugacci, Ulderico; Lavoué, Guillaume; Veltkamp, Remco C.
  • No Thumbnail Available
    Item
    SHREC 2023: Detection of Symmetries on 3D Point Clouds Representing Simple Shapes
    (The Eurographics Association, 2023) Sipiran, Ivan; Romanengo, Chiara; Falcidieno, Bianca; Biasotti, Silvia; Arvanitis, Gerasimos; Chen, Chen; Fotis, Vlassis; He, Jianfang; Lv, Xiaoling; Moustakas, Konstantinos; Peng, Silong; Romanelis, Ioannis; Sun, Wenhao; Vlachos, Christoforos; Wu, Ziyu; Xie, Qiong; Fugacci, Ulderico; Lavoué, Guillaume; Veltkamp, Remco C.
    This paper presents the methods that participated in the SHREC 2023 track focused on detecting symmetries on 3D point clouds representing simple shapes. By simple shapes, we mean surfaces generated by different types of closed plane curves used as the directrix of a cylinder or a cone. This track aims to determine the reflective planes for each point cloud. The methods are evaluated in their capability of detecting the right number of symmetries and correctly identifying the reflective planes. To this end, we generated a dataset that contains point clouds representing simple shapes perturbed with different kinds of artefacts (such as noise and undersampling) to provide a thorough evaluation of the robustness of the algorithms.
  • No Thumbnail Available
    Item
    synScan - A Large-Scale Dataset for Instance Level Recognition on Partial Scan Data
    (The Eurographics Association, 2023) Bookhahn, Marian; Neumann, Frank; Fugacci, Ulderico; Lavoué, Guillaume; Veltkamp, Remco C.
    Devices supporting depth-sensing technologies become more and more available, making it easier to access geometry-data driven services like 3D model or scene reconstruction. Utilizing these depth sensors, very large datasets have been created to enable deep learning for object detection and depth upsampling. We want to tackle the task of instance level recognition (ILR), where 3D scans of objects can be searched against a database of CAD models based on embeddings of their geometry. The distinctive property of this retrieval task is the existence of only a single corresponding database shape for each query. To the best of our knowledge all the existing datasets either lack in providing the exact CAD model correspondences or lack in scale and a variety of object categories. Therefore, we introduce synScan, a large-scale dataset synthetically generated via physically plausible domain randomization (PPDR) of 3D scenes and object-centric scan trajectories with the goal to mimic real-world object scan scenarios with a variety of incomplete views and occlusions. We provide approximately 39,000 randomly sampled scenes, made from 9,400 different shapes with semantic per-point annotation. We train and test different ILR algorithms (e.g. PointNetVLAD, MinkLoc3Dv2) designed for place-recognition in self-driving cars on the dataset and validate our results on a smaller real-world dataset. Utilizing a rather simple data generation pipeline, we can show that deep learning methods trained on our synthetic dataset can successfully adapt to real-world scan data. In this manner, synScan helps to overcome the lack of labeled training data.
  • No Thumbnail Available
    Item
    VariGrad: A Novel Feature Vector Architecture for Geometric Deep Learning on Unregistered Data
    (The Eurographics Association, 2023) Hartman, Emmanuel; Pierson, Emery; Fugacci, Ulderico; Lavoué, Guillaume; Veltkamp, Remco C.
    We present a novel geometric deep learning layer that leverages the varifold gradient (VariGrad) to compute feature vector representations of 3D geometric data. These feature vectors can be used in a variety of downstream learning tasks such as classification, registration, and shape reconstruction. Our model's use of parameterization independent varifold representations of geometric data allows our model to be both trained and tested on data independent of the given sampling or parameterization. We demonstrate the efficiency, generalizability, and robustness to resampling demonstrated by the proposed VariGrad layer.