EpicFlow: Edge-Preserving Interpolation of Correspondences for Optical Flow

Jerome Revaud 1 Philippe Weinzaepfel 1 Zaid Harchaoui 1 Cordelia Schmid 1
1 LEAR - Learning and recognition in vision
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
Abstract : We propose a novel approach for optical flow estima-tion, targeted at large displacements with significant oc-clusions. It consists of two steps: i) dense matching by edge-preserving interpolation from a sparse set of matches; ii) variational energy minimization initialized with the dense matches. The sparse-to-dense interpolation relies on an appropriate choice of the distance, namely an edge-aware geodesic distance. This distance is tailored to han-dle occlusions and motion boundaries, two common and difficult issues for optical flow computation. We also pro-pose an approximation scheme for the geodesic distance to allow fast computation without loss of performance. Sub-sequent to the dense interpolation step, standard one-level variational energy minimization is carried out on the dense matches to obtain the final flow estimation. The proposed approach, called Edge-Preserving Interpolation of Corre-spondences (EpicFlow) is fast and robust to large displace-ments. It significantly outperforms the state of the art on MPI-Sintel and performs on par on KITTI and Middlebury.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

Contributor : Thoth Team <>
Submitted on : Tuesday, January 13, 2015 - 4:53:32 PM
Last modification on : Monday, December 17, 2018 - 11:22:02 AM
Long-term archiving on: Tuesday, April 14, 2015 - 11:21:22 AM


Files produced by the author(s)


  • HAL Id : hal-01097477, version 1
  • ARXIV : 1501.02565


Jerome Revaud, Philippe Weinzaepfel, Zaid Harchaoui, Cordelia Schmid. EpicFlow: Edge-Preserving Interpolation of Correspondences for Optical Flow. 2015. ⟨hal-01097477v1⟩



Record views


Files downloads