Similarity propagation for one-shot and few-shot image segmentation
US11367271B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Jun 19, 2020 |
| Grant date | Jun 21, 2022 |
| Priority date | — |
| Expiry date | Nov 24, 2040 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06V10/761
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Embodiments of the present invention provide systems, methods, and computer storage media for one-shot and few-shot image segmentation on classes of objects that were not represented during training. In some embodiments, a dual prediction scheme may be applied in which query and support masks are jointly predicted using a shared decoder, which aids in similarity propagation between the query and support features. Additionally or alternatively, foreground and background attentive fusion may be applied to utilize cues from foreground and background feature similarities between the query and support images. Finally, to prevent overfitting on class-conditional similarities across training classes, input channel averaging may be applied for the query image during training. Accordingly, the techniques described herein may be used to achieve state-of-the-art performance for both one-shot and few-shot segmentation tasks.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.