Invention Grant
- Patent Title: Generating confidence-adaptive pixel-level predictions utilizing a multi-exit pixel-level prediction neural network
-
Application No.: US17214365Application Date: 2021-03-26
-
Publication No.: US11763545B2Publication Date: 2023-09-19
- Inventor: Evan Shelhamer , Zhuang Liu
- 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: G06V10/75
- IPC: G06V10/75 ; G06T3/40 ; G06V10/44 ; G06F18/213 ; G06F18/2411 ; G06N3/045

Abstract:
The present disclosure relates to systems, methods, and non-transitory computer readable media for efficiently, quickly, and flexibly generating and providing pixel-wise classification predictions utilizing early exit heads of a multi-exit pixel-level prediction neural network. For example, the disclosed systems utilize a multi-exit pixel-level prediction neural network to generate classification predictions for a digital image on the pixel level. The multi-exit pixel-level prediction neural network includes a specialized architecture with early exit heads having unique encoder-decoder architectures for generating pixel-wise classification predictions at different early exit stages. In some embodiments, the disclosed systems implement a spatial confidence-adaptive scheme to mask certain predicted pixels to prevent further processing of the masked pixels and thereby reduce computation.
Public/Granted literature
Information query