Title:
METHOD AND APPARATUS FOR SCENE SEGMENTATION FOR THREE-DIMENSIONAL SCENE RECONSTRUCTION
Document Type and Number:
WIPO Patent Application WO/2023/043061
Kind Code:
A1
Abstract:
A method includes obtaining, from an image sensor, image data of a real-world scene; obtaining, from a depth sensor, sparse depth data of the real-world scene; and passing the image data to a first neural network to obtain one or more object regions of interest (ROIs) and one or more feature map ROIs. Each object ROI includes at least one detected object. The method also includes passing the image data and sparse depth data to a second neural network to obtain one or more dense depth map ROIs; aligning the one or more object ROIs, one or more feature map ROIs, and one or more dense depth map ROIs; and passing the aligned ROIs to a fully convolutional network to obtain a segmentation of the real-world scene. The segmentation contains one or more pixelwise predictions of one or more objects in the real-world scene.
Inventors:
XIONG YINGEN (US)
PERI CHRISTOPHER A (US)
PERI CHRISTOPHER A (US)
Application Number:
PCT/KR2022/011777
Publication Date:
March 23, 2023
Filing Date:
August 08, 2022
Export Citation:
Assignee:
SAMSUNG ELECTRONICS CO LTD (KR)
International Classes:
G06T7/593; G06T7/10; G06T7/30; G06V10/25; G06V20/00
Foreign References:
| US20210174513A1 | 2021-06-10 | |||
| US20190037150A1 | 2019-01-31 | |||
| US20190147245A1 | 2019-05-16 | |||
| US20210073997A1 | 2021-03-11 | |||
| US20210183083A1 | 2021-06-17 |
Attorney, Agent or Firm:
Y.P.LEE, MOCK & PARTNERS (KR)
Download PDF: