Deep Parametric Continuous Convolutional Neural Networks

被引:2218
|
作者
Wang, Shenlong [1 ,3 ]
Suo, Simon [2 ,3 ]
Ma, Wei-Chiu [3 ]
Pokrovsky, Andrei [3 ]
Urtasun, Raquel [1 ,3 ]
机构
[1] Univ Toronto, Toronto, ON, Canada
[2] Univ Waterloo, Waterloo, ON, Canada
[3] Uber Adv Technol Grp, Pittsburgh, PA 15201 USA
关键词
D O I
10.1109/CVPR.2018.00274
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Standard convolutional neural networks assume a grid structured input is available and exploit discrete convolutions as their fundamental building blocks. This limits their applicability to many real-world applications. In this paper we propose Parametric Continuous Convolution, a new learnable operator that operates over non-grid structured data. The key idea is to exploit parameterized kernel functions that span the full continuous vector space. This generalization allows us to learn over arbitrary data structures as long as their support relationship is computable. Our experiments show significant improvement over the state-of-the-art in point cloud segmentation of indoor and outdoor scenes, and lidar motion estimation of driving scenes.
引用
收藏
页码:2589 / 2597
页数:9
相关论文
共 50 条
  • [31] Deep Convolutional Neural Networks for DGA Detection
    Catania, Carlos
    Garcia, Sebastian
    Torres, Pablo
    [J]. COMPUTER SCIENCE - CACIC 2018, 2019, 995 : 327 - 340
  • [32] Energy Propagation in Deep Convolutional Neural Networks
    Wiatowski, Thomas
    Grohs, Philipp
    Boelcskei, Helmut
    [J]. IEEE TRANSACTIONS ON INFORMATION THEORY, 2018, 64 (07) : 4819 - 4842
  • [33] Predicting enhancers with deep convolutional neural networks
    Xu Min
    Wanwen Zeng
    Shengquan Chen
    Ning Chen
    Ting Chen
    Rui Jiang
    [J]. BMC Bioinformatics, 18
  • [34] Elastography mapped by deep convolutional neural networks
    LIU DongXu
    KRUGGEL Frithjof
    SUN LiZhi
    [J]. Science China Technological Sciences, 2021, 64 (07) : 1567 - 1574
  • [35] WEATHER CLASSIFICATION WITH DEEP CONVOLUTIONAL NEURAL NETWORKS
    Elhoseiny, Mohamed
    Huang, Sheng
    Elgammal, Ahmed
    [J]. 2015 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2015, : 3349 - 3353
  • [36] Universal Consistency of Deep Convolutional Neural Networks
    Lin, Shao-Bo
    Wang, Kaidong
    Wang, Yao
    Zhou, Ding-Xuan
    [J]. IEEE TRANSACTIONS ON INFORMATION THEORY, 2022, 68 (07) : 4610 - 4617
  • [37] Deep convolutional neural networks in the face of caricature
    Hill, Matthew Q.
    Parde, Connor J.
    Castillo, Carlos D.
    Colon, Y. Ivette
    Ranjan, Rajeev
    Chen, Jun-Cheng
    Blanz, Volker
    O'Toole, Alice J.
    [J]. NATURE MACHINE INTELLIGENCE, 2019, 1 (11) : 522 - 529
  • [38] Refining Architectures of Deep Convolutional Neural Networks
    Shankar, Sukrit
    Robertson, Duncan
    Ioannou, Yani
    Criminisi, Antonio
    Cipolla, Roberto
    [J]. 2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, : 2212 - 2220
  • [39] Very Deep Convolutional Neural Networks for LVCSR
    Bi, Mengxiao
    Qian, Yanmin
    Yu, Kai
    [J]. 16TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2015), VOLS 1-5, 2015, : 3259 - 3263
  • [40] Plankton Classification with Deep Convolutional Neural Networks
    Ouyang Py
    Hu Hong
    Shi Zhongzhi
    [J]. 2016 IEEE INFORMATION TECHNOLOGY, NETWORKING, ELECTRONIC AND AUTOMATION CONTROL CONFERENCE (ITNEC), 2016, : 132 - 136