English

Intelligent Autofocus

Image and Video Processing 2020-03-02 v1

Abstract

We demonstrate that deep learning methods can determine the best focus position from 1-2 image samples, enabling 5-10x faster focus than traditional search-based methods. In contrast with phase detection methods, deep autofocus does not require specialized hardware. In further constrast with conventional methods, which assume a static "best focus," AI methods can generate scene-based focus trajectories that optimize synthesized image quality for dynamic and three dimensional scenes.

Keywords

Cite

@article{arxiv.2002.12389,
  title  = {Intelligent Autofocus},
  author = {Chengyu Wang and Qian Huang and Ming Cheng and Zhan Ma and David J. Brady},
  journal= {arXiv preprint arXiv:2002.12389},
  year   = {2020}
}
R2 v1 2026-06-23T13:56:47.550Z