DeepMatching: Hierarchical Deformable Dense Matching

  title={DeepMatching: Hierarchical Deformable Dense Matching},
  author={J{\'e}r{\^o}me Revaud and Philippe Weinzaepfel and Z. Harchaoui and C. Schmid},
  journal={International Journal of Computer Vision},
  • Jérôme Revaud, Philippe Weinzaepfel, +1 author C. Schmid
  • Published 2016
  • Computer Science
  • International Journal of Computer Vision
  • We introduce a novel matching algorithm, called DeepMatching, to compute dense correspondences between images. [...] Key Method We also apply DeepMatching to the computation of optical flow, called DeepFlow, by integrating it in the large displacement optical flow (LDOF) approach of Brox and Malik (Large displacement optical flow: descriptor matching in variational motion estimation, 2011). Additional robustness to large displacements and complex motion is obtained thanks to our matching approach. DeepFlow…Expand Abstract
    Universal Correspondence Network
    • 170
    • Highly Influenced
    • Open Access
    Locality Sensitive Hashing based deepmatching for optical flow estimation
    Deep Semantic Feature Matching Using Confidential Correspondence Consistency
    Learning Semantic Correspondence Exploiting an Object-level Prior
    • 1
    • Open Access
    Deep Semantic Feature Matching
    • 40
    • Open Access
    Boosting Local Matches with Convolutional Co-Segmentation
    SFNet: Learning Object-Aware Semantic Correspondence
    • 10
    • Highly Influenced
    • Open Access
    Multi-Scale Generalized Plane Match for Optical Flow


    Publications referenced by this paper.
    Non-rigid dense correspondence with applications for image enhancement
    • 105
    • Highly Influential
    Locally Affine Sparse-to-Dense Matching for Motion and Occlusion Estimation
    • 68
    • Open Access
    Deformable Spatial Pyramid Matching for Fast Dense Correspondences
    • 221
    • Highly Influential
    • Open Access
    DAISY Filter Flow: A Generalized Discrete Approach to Dense Correspondences
    • 77
    • Highly Influential
    • Open Access
    SIFT Flow: Dense Correspondence across Scenes and Its Applications
    • 1,330
    • Highly Influential
    • Open Access
    A General Dense Image Matching Framework Combining Direct and Feature-Based Costs
    • 56
    • Open Access
    Adaptive Integration of Feature Matches into Variational Optical Flow Methods
    • 30
    • Open Access