Patent · US Active

System and method for boundary aware semantic segmentation

US11461998B2 · kind B2 · utility

1Cited by

3References

18Claims

0Family size

Assignee

SAMSUNG ELECTRONICS CO., LTD. · KR

Inventors

Qingfeng Liu · 东风镇, CN
Mostafa El-Khamy · San Diego, US
Dongwoon Bai · San Diego, US
Jungwon Lee · Seoul, KR

Key dates

Filing date	Jan 30, 2020
Grant date	Oct 4, 2022
Priority date	—
Expiry date	Jan 20, 2041

Classification

Technology area (CPC G)Physics
CPC primaryG06V20/70
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

Some aspects of embodiments of the present disclosure relate to using a boundary aware loss function to train a machine learning model for computing semantic segmentation maps from input images. Some aspects of embodiments of the present disclosure relate to deep convolutional neural networks (DCNNs) for computing semantic segmentation maps from input images, where the DCNNs include a box filtering layer configured to box filter input feature maps computed from the input images before supplying box filtered feature maps to an atrous spatial pyramidal pooling (ASPP) layer. Some aspects of embodiments of the present disclosure relate to a selective ASPP layer configured to weight the outputs of an ASPP layer in accordance with attention feature maps.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.