Generating digital elevation models from satellite imagery using neural radiance field

HIGHLIGHTS

• What: The authors propose an enhanced method based on the Sat-NeRF model to generate 3D models from a set of multiview satellite images of a scene. The experiments were conducted using the Data Fusion Contest (DFC2019) dataset, which includes features of Jacksonville (JAX) obtained from satellite images captured by WorldView-3, with a resolution of 0.3 m per pixel, over the course of a year. Compared with existing methods, the experiments demonstrate that the approach can produce high-quality DEMs and corresponding mesh models.
• Who: DEM et al. from the School of Geomatics . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or Email

  Password

  Remember me

  Lost your password?

  If you don't have an account, you can create one here.