ML Anthology

Aperture Supervision for Monocular Depth Estimation

Pratul P. Srinivasan, Rahul Garg, Neal Wadhwa, Ren Ng, Jonathan T. Barron
CVPR 2018
doi:10.1109/CVPR.2018.00669 /cvpr/2018/srinivasan2018cvpr-aperture/

Abstract

We present a novel method to train machine learning algorithms to estimate scene depths from a single image, by using the information provided by a camera's aperture as supervision. Prior works use a depth sensor's outputs or images of the same scene from alternate viewpoints as supervision, while our method instead uses images from the same viewpoint taken with a varying camera aperture. To enable learning algorithms to use aperture effects as supervision, we introduce two differentiable aperture rendering functions that use the input image and predicted depths to simulate the depth-of-field effects caused by real camera apertures. We train a monocular depth estimation network end-to-end to predict the scene depths that best explain these finite aperture images as defocus-blurred renderings of the input all-in-focus image.

PDF CVPR Semantic Scholar

Cite

Text

Srinivasan et al. "Aperture Supervision for Monocular Depth Estimation." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018. doi:10.1109/CVPR.2018.00669

Markdown

[Srinivasan et al. "Aperture Supervision for Monocular Depth Estimation." Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018.](https://mlanthology.org/cvpr/2018/srinivasan2018cvpr-aperture/) doi:10.1109/CVPR.2018.00669

BibTeX

@inproceedings{srinivasan2018cvpr-aperture,
  title     = {{Aperture Supervision for Monocular Depth Estimation}},
  author    = {Srinivasan, Pratul P. and Garg, Rahul and Wadhwa, Neal and Ng, Ren and Barron, Jonathan T.},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year      = {2018},
  doi       = {10.1109/CVPR.2018.00669},
  url       = {https://mlanthology.org/cvpr/2018/srinivasan2018cvpr-aperture/}
}