Depth and DOF Cues Make A Better Defocus Blur Detector

Yuxin Jin, Ming Qian, Jincheng Xiong, Nan Xue, Gui-Song Xia
6/20/2023
cs.CV

No official code found. Best implementation: yuxinjin-whu/D-DFFNet (45 stars, last commit 2023-09-14).

Abstract

Defocus blur detection (DBD) separates in-focus and out-of-focus regions in an image. Previous approaches mistakenly mistook homogeneous areas in focus for defocus blur regions, likely due to not considering the internal factors that cause defocus blur. Inspired by the law of depth, depth of field (DOF), and defocus, we propose an approach called D-DFFNet, which incorporates depth and DOF cues in an implicit manner. This allows the model to understand the defocus phenomenon in a more natural way. Our method proposes a depth feature distillation strategy to obtain depth knowledge from a pre-trained monocular depth estimation model and uses a DOF-edge loss to understand the relationship between DOF and depth. Our approach outperforms state-of-the-art methods on public benchmarks and a newly collected large benchmark dataset, EBD. Source codes and EBD dataset are available at: https:github.com/yuxinjin-whu/D-DFFNet.

View on arXivView PDF

Code Implementations(1)

yuxinjin-whu/D-DFFNetCommunity100%

Depth and DOF Cues Make A Better Defocus Blur Detector

455Shell, PythonMar 28, 20233 years ago

Similar Papers

Camera-Independent Single Image Depth Estimation from Defocus Blur

90% match2023

Lens Parameter Estimation for Realistic Depth of Field Modeling

87% match2026

Direct Handheld Burst Imaging to Simulated Defocus

87% match2022

CoCoGaussian: Leveraging Circle of Confusion for Gaussian Splatting from Defocused Images

86% match1 repo462026

Depth Estimation and Image Restoration by Deep Learning from Defocused Images

85% match1 repo52023

A Novel Defocus-Blur Region Detection Approach Based on DCT Feature and PCNN Structure

85% match2023

Cite this paper

@article{jin2023depth,
  title  = {Depth and DOF Cues Make A Better Defocus Blur Detector},
  author = {Yuxin Jin and Ming Qian and Jincheng Xiong and Nan Xue and Gui-Song Xia},
  year   = {2023},
  eprint = {2306.11334},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2306.11334v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2306.11334.svg)](https://codeofpaper.com/paper/2306.11334)

Discussion