US 11,967,096 B2
Methods and apparatuses of depth estimation from focus information
Ren Wang, Hsinchu (TW); Yu-Lun Liu, Hsinchu (TW); Yu-Hao Huang, Hsinchu (TW); and Ning-Hsu Wang, Hsinchu (TW)
Assigned to MEDIATEK INC., Hsinchu (TW)
Filed by MEDIATEK INC., Hsinchu (TW)
Filed on Feb. 22, 2022, as Appl. No. 17/677,365.
Claims priority of provisional application 63/200,692, filed on Mar. 23, 2021.
Prior Publication US 2022/0309696 A1, Sep. 29, 2022
Int. Cl. G06T 7/571 (2017.01); G06N 20/00 (2019.01); G06T 7/557 (2017.01); H04N 5/222 (2006.01); H04N 23/67 (2023.01)
CPC G06T 7/571 (2017.01) [G06N 20/00 (2019.01); G06T 7/557 (2017.01); H04N 5/2226 (2013.01); H04N 23/67 (2023.01)] 19 Claims
OG exemplary drawing
 
1. A depth estimation from focus method for an image or video processing system, comprising:
receiving input image data containing focus information;
generating an intermediate attention map from the input image data containing focus information by an Artificial Intelligence (AI) model, wherein the AI model is trained by one or more focus-based training materials;
normalizing the intermediate attention map into a depth attention map via a first normalization function; and
deriving expected depth values for the input image data containing focus information from the depth attention map.