Invention Grant
- Patent Title: Generating depth images utilizing a machine-learning model built from mixed digital image sources and multiple loss function sets
-
Application No.: US17186436Application Date: 2021-02-26
-
Publication No.: US11798180B2Publication Date: 2023-10-24
- Inventor: Wei Yin , Jianming Zhang , Oliver Wang , Simon Niklaus , Mai Long , Su Chen
- Applicant: Adobe Inc.
- Applicant Address: US CA San Jose
- Assignee: Adobe Inc.
- Current Assignee: Adobe Inc.
- Current Assignee Address: US CA San Jose
- Agency: Keller Preece PLLC
- Main IPC: G06T7/50
- IPC: G06T7/50 ; G06T7/593 ; G06T7/30 ; G06T7/13 ; G06T7/143 ; G06T7/521

Abstract:
This disclosure describes one or more implementations of a depth prediction system that generates accurate depth images from single input digital images. In one or more implementations, the depth prediction system enforces different sets of loss functions across mix-data sources to generate a multi-branch architecture depth prediction model. For instance, in one or more implementations, the depth prediction model utilizes different data sources having different granularities of ground truth depth data to robustly train a depth prediction model. Further, given the different ground truth depth data granularities from the different data sources, the depth prediction model enforces different combinations of loss functions including an image-level normalized regression loss function and/or a pair-wise normal loss among other loss functions.
Public/Granted literature
Information query