3D SCENE RECONSTRUCTION FROM RGB IMAGES

被引:0
|
作者
Rotaru, Razvan-Paul [1 ]
Gradinaru, Alexandru [1 ]
Moldoveanu, Florica [1 ]
机构
[1] Natl Univ Sci & Technol POLITEHN Bucharest, Fac Comp Sci & Informat Technol, Bucharest, Romania
关键词
3D scene reconstruction; object detection; pose estimation;
D O I
暂无
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Object recognition is significantly improving, allowing us to better understand and extract information from images. This paper presents a novel method for 3D scene reconstruction using a single RGB image, based on a known 3D model database. We use every detected object in an image to further process its pose and a corresponding 3D model by leveraging existing datasets of both 3D models and labeled images to understand and simulate perception correctly. This method produces a clean and lightweight representation of a scene. State-of-the-art research, implementation details, and evaluation results are presented.
引用
收藏
页码:101 / 112
页数:12
相关论文
共 50 条
  • [1] 3D SCENE RECONSTRUCTION FROM RGB IMAGES
    Rotaru, Răzvan-Paul
    Grădinaru, Alexandru
    Moldoveanu, Florica
    UPB Scientific Bulletin, Series C: Electrical Engineering and Computer Science, 2024, 86 (02): : 101 - 112
  • [2] Panoptic 3D Scene Reconstruction From a Single RGB Image
    Dahnert, Manuel
    Hou, Ji
    Niessner, Matthias
    Dai, Angela
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 34 (NEURIPS 2021), 2021, 34
  • [3] 3D Scene Reconstruction from RGB Images Using Dynamic Graph Convolution for Augmented Reality
    Weng, Tzu-Hsuan
    Fischer, Robin
    Fu, Li-Chen
    2022 IEEE CONFERENCE ON VIRTUAL REALITY AND 3D USER INTERFACES ABSTRACTS AND WORKSHOPS (VRW 2022), 2022, : 629 - 630
  • [4] Iterative Online 3D Reconstruction from RGB Images
    Cardoen, Thorsten
    Leroux, Sam
    Simoens, Pieter
    SENSORS, 2022, 22 (24)
  • [5] Learning 3D Scene Synthesis from Annotated RGB-D Images
    Kermani, Z. Sadeghipour
    Liao, Z.
    Tan, P.
    Zhang, H.
    COMPUTER GRAPHICS FORUM, 2016, 35 (05) : 197 - 206
  • [6] 3D scene reconstruction from cylindrical panoramic images
    Bunschoten, R
    Kröse, B
    ROBOTICS AND AUTONOMOUS SYSTEMS, 2002, 41 (2-3) : 111 - 118
  • [7] Holistic 3D Scene Parsing and Reconstruction from a Single RGB Image
    Huang, Siyuan
    Qi, Siyuan
    Zhu, Yixin
    Xiao, Yinxue
    Xu, Yuanlu
    Zhu, Song-Chun
    COMPUTER VISION - ECCV 2018, PT VII, 2018, 11211 : 194 - 211
  • [8] 3D geometry reconstruction from gray and RGB medical images
    Talaia, P.
    Parente, M.
    Fernandes, A.
    Natal Jorge, R.
    COMPUTATIONAL VISION AND MEDICAL IMAGE PROCESSING: VIPIMAGE 2011, 2012, : 297 - 301
  • [9] 3D scene reconstruction from reflection images in a spherical mirror
    Kanbara, Masayuki
    Ukita, Norimichi
    Kidode, Masatsugu
    Yokoya, Naokazu
    18TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION, VOL 4, PROCEEDINGS, 2006, : 874 - +
  • [10] 3D Scene Reconstruction from Multi-aperture Images
    Mao, Miao
    Qin, Kaihuai
    6TH INTERNATIONAL CONFERENCE ON DIGITAL IMAGE PROCESSING (ICDIP 2014), 2014, 9159