公开(公告)号：US20240233339A1

公开(公告)日：2024-07-11

申请号：US18325182

申请日：2023-05-30

Applicant: ECOPIA TECH CORPORATION

Inventor： Yuanming SHU ,  Yaolong WANG ,  Hongbo WANG

IPC: G06V10/774 ,  G06V20/10

CPC classification number: G06V10/774 ,  G06V20/176

Abstract: Example systems and methods for data augmentation for occlusion handling in feature extraction are provided. An example method involves accessing a geospatial image depicting an occludable landcover feature; augmenting the geospatial image with a simulated occluding landcover feature to generate an occluding feature-augmented geospatial image, wherein the simulated occluding landcover feature partially occludes the occludable landcover feature as depicted in the geospatial image; and training a machine learning model, using the occluding feature-augmented geospatial image, to extract geometric representations of occludable landcover features as depicted in geospatial imagery.

公开(公告)号：US20240233261A1

公开(公告)日：2024-07-11

申请号：US18319553

申请日：2023-05-18

Applicant: ECOPIA TECH CORPORATION

Inventor： Yuanming SHU ,  Shuo TAN ,  Zihao CHEN ,  Ruijie DENG

IPC: G06T17/05 ,  G06T7/60 ,  G06T17/10 ,  G06V20/10

CPC classification number: G06T17/05 ,  G06T7/60 ,  G06T17/10 ,  G06V20/176 ,  G06T2210/04 ,  G06V2201/12

Abstract: Methods and systems for determining the height of structures based on imagery of the structures and associated two-dimensional vector data are provided. An example method involves projecting two-dimensional vector data outlining a roof of a structure into images of the structure captured from different perspectives and feature matching the vector data across the imagery to determine a best-matching three-dimensional position for the roof situated at the height of the structure.

公开(公告)号：US20230351728A1

公开(公告)日：2023-11-02

申请号：US17731769

申请日：2022-04-28

Applicant: ECOPIA TECH CORPORATION

Inventor： Yuanming SHU

IPC: G06V10/766 ,  G06V10/77 ,  G06V10/75 ,  G06V10/82 ,  G06N20/00 ,  G06V10/774

CPC classification number: G06V10/766 ,  G06V10/7715 ,  G06V10/75 ,  G06V10/82 ,  G06N20/00 ,  G06V10/774

Abstract: Methods and systems for generating vector maps representing features depicted in imagery are provided. An example method involves obtaining remote imagery that depicts a feature, applying a machine learning model to the remote imagery to extract a geometric model of the feature encoded as a tokenized representation of a sequence of annotation operations, and interpreting the tokenized representation as a vector map representing the geometric model of the feature.