Corpus ID: 12578058

Universal Correspondence Network

@inproceedings{Choy2016UniversalCN,
  title={Universal Correspondence Network},
  author={Christopher Bongsoo Choy and JunYoung Gwak and Silvio Savarese and Manmohan Krishna Chandraker},
  booktitle={NIPS},
  year={2016}
}
  • Christopher Bongsoo Choy, JunYoung Gwak, +1 author Manmohan Krishna Chandraker
  • Published in NIPS 2016
  • Computer Science
  • We present a deep learning framework for accurate visual correspondences and demonstrate its effectiveness for both geometric and semantic matching, spanning across rigid motions to intra-class shape or appearance variations. In contrast to previous CNN-based approaches that optimize a surrogate patch similarity objective, we use deep metric learning to directly learn a feature space that preserves either geometric or semantic similarity. Our fully convolutional architecture, along with a novel… CONTINUE READING

    Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 136 CITATIONS

    Semantic Correspondence in the Wild

    VIEW 14 EXCERPTS
    CITES RESULTS, BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Deep Learning of Graph Matching

    VIEW 6 EXCERPTS
    CITES METHODS, BACKGROUND & RESULTS
    HIGHLY INFLUENCED

    SCNet: Learning Semantic Correspondence

    VIEW 13 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Extremely Dense Point Correspondences using a Learned Feature Descriptor

    VIEW 12 EXCERPTS
    CITES METHODS, BACKGROUND & RESULTS
    HIGHLY INFLUENCED

    ELF: Embedded Localisation of Features in Pre-Trained CNN

    VIEW 7 EXCERPTS
    CITES METHODS, BACKGROUND & RESULTS
    HIGHLY INFLUENCED

    Geometric Correspondence Network for Camera Motion Estimation

    VIEW 6 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Object Pose Estimation from Monocular Image using Multi-View Keypoint Correspondence

    VIEW 6 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Recurrent Transformer Networks for Semantic Correspondence

    VIEW 13 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Semantic Correspondence: A Hierarchical Approach

    VIEW 12 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    FILTER CITATIONS BY YEAR

    2016
    2020

    CITATION STATISTICS

    • 25 Highly Influenced Citations

    • Averaged 39 Citations per year from 2017 through 2019

    • 56% Increase in citations per year in 2019 over 2018

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 36 REFERENCES

    Spatial Transformer Networks

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    DeepMatching: Hierarchical Deformable Dense Matching

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    WarpNet: Weakly Supervised Matching for Single-View Reconstruction

    VIEW 7 EXCERPTS

    Computing the stereo matching cost with a convolutional neural network

    • Jure Zbontar, Yann LeCun
    • Computer Science
    • 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
    • 2015
    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    FlowWeb: Joint image set alignment by weaving consistent, pixel-wise correspondences

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Learning to See by Moving

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Learning to compare image patches via convolutional neural networks

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Do Convnets Learn Correspondence?

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Caltech-UCSD Birds 200

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL