Abstract

We present a novel, to the best of our knowledge, patch-based approach for depth regression from defocus blur. Most state-of-the-art methods for depth from defocus (DFD) use a patch classification approach among a set of potential defocus blurs related to a depth, which induces errors due to the continuous variation of the depth. Here, we propose to adapt a simple classification model using a soft-assignment encoding of the true depth into a membership probability vector during training and a regression scale to predict intermediate depth values. Our method uses no blur model or scene model; it only requires a training dataset of image patches (either raw, gray scale, or RGB) and their corresponding depth label. We show that our method outperforms both classification and direct regression on simulated images from structured or natural texture datasets, and on raw real data having optical aberrations from an active DFD experiment.

© 2022 Optica Publishing Group

Learning scene and blur model for active chromatic depth from defocus

Benjamin Buat, Pauline Trouvé-Peloux, Frédéric Champagnat, and Guy Le Besnerais
Appl. Opt. 60(31) 9966-9974 (2021)

Passive depth estimation using chromatic aberration and a depth from defocus approach

Pauline Trouvé, Frédéric Champagnat, Guy Le Besnerais, Jacques Sabater, Thierry Avignon, and Jérôme Idier
Appl. Opt. 52(29) 7152-7164 (2013)

Rational-operator-based depth-from-defocus approach to scene reconstruction

Ang Li, Richard Staunton, and Tardi Tjahjadi
J. Opt. Soc. Am. A 30(9) 1787-1795 (2013)

Performance model of depth from defocus with an unconventional camera

P. Trouvé-Peloux, F. Champagnat, G. Le Besnerais, G. Druart, and J. Idier
J. Opt. Soc. Am. A 38(10) 1489-1500 (2021)

Depth estimation using spectrally varying defocus blur

Shin Ishihara, Antonin Sulc, and Imari Sato
J. Opt. Soc. Am. A 38(8) 1140-1149 (2021)