We decouple the geometry and texture of 2DGS,
representing the appearance with a global neural texture while using 2DGS as a geometric sampler.
This approach reconstructs richer texture details while effectively reducing the number of primitives.
This paper aims to develop an accurate 3D geometry representation of satellite images using
satellite-ground image pairs. Our focus is on the challenging problem of generating ground-view
panoramas from satellite images.
Honors
Outstanding Graduate of Wuhan University 2024
Postgraduate Recommendation of Wuhan University (1/198) 2023
Jun Lei Scholarship, Xiaomi Inc. 2023
Outstanding Student of Wuhan University 2021, 2023
First/Second Class Scholarship of Wuhan University 2021, 2022, 2023
The page template is referenced from Jon Barron's website.