ScopeFlow: Dynamic Scene Scoping for Optical Flow

CVPR 2020  ·  Aviram Bar-Haim, Lior Wolf ·

We propose to modify the common training protocols of optical flow, leading to sizable accuracy improvements without adding to the computational complexity of the training process. The improvement is based on observing the bias in sampling challenging data that exists in the current training protocol, and improving the sampling process. In addition, we find that both regularization and augmentation should decrease during the training protocol. Using an existing low parameters architecture, the method is ranked first on the MPI Sintel benchmark among all other methods, improving the best two frames method accuracy by more than 10%. The method also surpasses all similar architecture variants by more than 12% and 19.7% on the KITTI benchmarks, achieving the lowest Average End-Point Error on KITTI2012 among two-frame methods, without using extra datasets.

PDF Abstract CVPR 2020 PDF CVPR 2020 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Optical Flow Estimation Sintel-clean ScopeFlow Average End-Point Error 3.592 # 19
Optical Flow Estimation Sintel-final ScopeFlow Average End-Point Error 4.098 # 12

Methods


No methods listed for this paper. Add relevant methods here