Depth Completion using Geometry-Aware Embedding

被引:21
|
作者
Du, Wenchao [1 ]
Chen, Hu [1 ]
Yang, Hongyu [1 ]
Zhang, Yi [1 ]
机构
[1] Sichuan Univ, Coll Comp Sci, Chengdu 610065, Peoples R China
关键词
NETWORK; SPARSE;
D O I
10.1109/ICRA46639.2022.9811556
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Exploiting internal spatial geometric constraints of sparse LiDARs is beneficial to depth completion, however, has been not explored well. This paper proposes an efficient method to learn geometry-aware embedding, which encodes the local and global geometric structure information from 3D points, e.g., scene layout, object's sizes and shapes, to guide dense depth estimation. Specifically, we utilize the dynamic graph representation to model generalized geometric relationship from irregular point clouds in a flexible and efficient manner. Further, we joint this embedding and corresponded RGB appearance information to infer missing depths of the scene with well structure-preserved details. The key to our method is to integrate implicit 3D geometric representation into a 2D learning architecture, which leads to a better trade-off between the performance and efficiency. Extensive experiments demonstrate that the proposed method outperforms previous works and could reconstruct fine depths with crisp boundaries in regions that are over-smoothed by them. The ablation study gives more insights into our method that could achieve significant gains with a simple design, while having better generalization capability and stability. The code is available at https://github.com/Wenchao-Du/GAENet.
引用
收藏
页码:8680 / 8686
页数:7
相关论文
共 50 条
  • [21] Geometry-Aware Adaptation for Pretrained Models
    Roberts, Nicholas
    Li, Xintong
    Adila, Dyah
    Cromp, Sonia
    Huang, Tzu-Heng
    Zhao, Jitian
    Sala, Frederic
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 36 (NEURIPS 2023), 2023,
  • [22] GAST: Geometry-Aware Structure Transformer
    Khomiakov, Maxim
    Andersen, Michael Riis
    Frellsen, Jes
    2024 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION WORKSHOPS, WACVW 2024, 2024, : 776 - 784
  • [23] Multi-view Relighting using a Geometry-Aware Network
    Philip, Julien
    Gharbi, Michael
    Zhou, Tinghui
    Efros, Alexei A.
    Drettakis, George
    ACM TRANSACTIONS ON GRAPHICS, 2019, 38 (04):
  • [24] Geometry-Aware Metropolis Light Transport
    Otsu, Hisanari
    Hanika, Johannes
    Hachisuka, Toshiya
    Dachsbacher, Carsten
    ACM TRANSACTIONS ON GRAPHICS, 2018, 37 (06):
  • [25] Geometry-Aware Metropolis Light Transport
    Otsu, Hisanari
    Hanika, Johannes
    Hachisuka, Toshiya
    Dachsbacher, Carsten
    SIGGRAPH ASIA'18: SIGGRAPH ASIA 2018 TECHNICAL PAPERS, 2018,
  • [26] Smoothing of meshes and point clouds using weighted geometry-aware bases
    Volodine, Tim
    Vanderstraeten, Denis
    Roose, Dirk
    GEOMETRIC MODELING AND PROCESSING - GMP 2006, PROCEEDINGS, 2006, 4077 : 687 - 693
  • [27] Learning Geometry-Aware Kernels in a Regularization Framework
    Pan, Binbin
    Chen, Wen-Sheng
    COMPUTER ANALYSIS OF IMAGES AND PATTERNS, PT I, 2013, 8047 : 352 - 359
  • [28] Geometry-aware manipulability learning, tracking, and transfer
    Jaquier, Noemie
    Rozo, Leonel
    Caldwell, Darwin G.
    Calinon, Sylvain
    INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH, 2021, 40 (2-3): : 624 - 650
  • [29] Geometry-Aware Hierarchical Bayesian Learning on Manifolds
    Fan, Yonghui
    Wang, Yalin
    2022 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION (WACV 2022), 2022, : 2743 - 2752
  • [30] Plant point cloud completion network based on multi-scale geometry-aware point Transformer
    Zeng A.
    Peng J.
    Liu C.
    Pan D.
    Jiang Y.
    Zhang X.
    Nongye Gongcheng Xuebao/Transactions of the Chinese Society of Agricultural Engineering, 2022, 38 (04): : 198 - 205