Surface Normals in the Wild

ICCV 2017

doi:10.1109/ICCV.2017.173 /iccv/2017/chen2017iccv-surface/

Abstract

We study the problem of single-image depth estimation for images in the wild. We collect human annotated surface normals and use them to help train a neural network that directly predicts pixel-wise depth. We propose two novel loss functions for training with surface normal annotations. Experiments on NYU Depth, KITTI, and our own dataset demonstrate that our approach can significantly improve the quality of depth estimation in the wild.

PDF ICCV Semantic Scholar

Cite

Text

Chen et al. "Surface Normals in the Wild." International Conference on Computer Vision, 2017. doi:10.1109/ICCV.2017.173

Markdown

[Chen et al. "Surface Normals in the Wild." International Conference on Computer Vision, 2017.](https://mlanthology.org/iccv/2017/chen2017iccv-surface/) doi:10.1109/ICCV.2017.173

BibTeX

@inproceedings{chen2017iccv-surface,
  title     = {{Surface Normals in the Wild}},
  author    = {Chen, Weifeng and Xiang, Donglai and Deng, Jia},
  booktitle = {International Conference on Computer Vision},
  year      = {2017},
  doi       = {10.1109/ICCV.2017.173},
  url       = {https://mlanthology.org/iccv/2017/chen2017iccv-surface/}
}