Boundary Flow: A Siamese Network that Predicts Boundary Motion Without Training on Motion

@article{Lei2018BoundaryFA,
  title={Boundary Flow: A Siamese Network that Predicts Boundary Motion Without Training on Motion},
  author={P. Lei and F. Li and S. Todorovic},
  journal={2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year={2018},
  pages={3282-3290}
}
  • P. Lei, F. Li, S. Todorovic
  • Published 2018
  • Computer Science
  • 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
  • Using deep learning, this paper addresses the problem of joint object boundary detection and boundary motion estimation in videos, which we named boundary flow estimation. [...] Key Method For boundary flow estimation, we specify a new fully convolutional Siamese network (FCSN) that jointly estimates object-level boundaries in two consecutive frames. Boundary correspondences in the two frames are predicted by the same FCSN with a new, unconventional deconvolution approach.Expand Abstract
    AN ABSTRACT OF THE DISSERTATION OF
    Object Contour and Edge Detection with RefineContourNet
    2
    StableNet: Semi-Online, Multi-Scale Deep Video Stabilization
    1
    Leveraging Shape Completion for 3D Siamese Tracking
    15
    REDN: A Recursive Encoder-Decoder Network for Edge Detection
    1
    Modular Autonomous Taxiing Simulation and 3D Siamese Vehicle Tracking

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 47 REFERENCES
    Exploiting Discontinuities in Optical Flow
    30
    DeepFlow: Large Displacement Optical Flow with Deep Matching
    750
    EpicFlow: Edge-preserving interpolation of correspondences for optical flow
    564
    Key-segments for video object segmentation
    453
    Object Contour Detection with a Fully Convolutional Encoder-Decoder Network
    197
    Dynamic Occlusion Analysis in Optical Flow Fields
    64
    Holistically-Nested Edge Detection
    1392
    Efficient Coarse-to-Fine Patch Match for Large Displacement Optical Flow
    143