Home
Uni-Logo
 

FusionNet and AugmentedFlowNet: Selective Proxy Ground Truth for Training on Unlabeled Images

Technical Report , arXiv:1808.06389, 2018
Abstract: Recent work has shown that convolutional neural networks (CNNs) can be used to estimate optical flow with high quality and fast runtime. This makes them preferable for real-world applications. However, such networks require very large training datasets. Engineering the training data is difficult and/or laborious. This paper shows how to augment a network trained on an existing synthetic dataset with large amounts of additional unlabelled data. In particular, we introduce a selection mechanism to assemble from multiple estimates a joint optical flow field, which outperforms that of all input methods. The latter can be used as proxy-ground-truth to train a network on real-world data and to adapt it to specific domains of interest. Our experimental results show that the performance of networks improves considerably, both, in cross-domain and in domain-specific scenarios. As a consequence, we obtain state-of-the-art results on the KITTI benchmarks.
Paper

Images and movies

 

BibTex reference

@TechReport{MIB18,
  author       = "O. Makansi and E. Ilg and T. Brox",
  title        = "FusionNet and AugmentedFlowNet: Selective Proxy Ground Truth for Training on Unlabeled Images",
  institution  = "arXiv:1808.06389",
  month        = " ",
  year         = "2018",
  url          = "http://lmb.informatik.uni-freiburg.de/Publications/2018/MIB18"
}

Other publications in the database