Using AI to Improve 3D Reconstruction from Remote Sensing Imagery

Event Time

Originally Aired - Tuesday, May 7 2:00 PM - 3:00 PM

Info Alert

Create or Log in to My Show Planner to see Videos and Resources.

Info Alert

Your account does not have access to this session.

Videos

Resources


{{video.title}}

Log in to your planner to join the zoom meeting!

{{chatHeaderContent}}

{{chatBodyContent}}

Session Recording

Info Alert

This Session Has Not Started Yet

Be sure to come back after the session starts to have access to session resources.

Event Location

Location: Sun Ballroom 3-4


Event Information

Title: Using AI to Improve 3D Reconstruction from Remote Sensing Imagery

Description:

Training Summary: In the last few years the methods used for reconstruction of 3D geometry from imagery have undergone an AI revolution. These new methods are now being adapted to remote sensing applications as well. This training session will review recent developments in the computer vision and AI research communities as they pertain to 3D reconstruction from overhead imagery. We will provide a high-level summary of how traditional multi-view 3D reconstruction works, review the basics of neural networks, and then discuss how these fields have come together by reviewing and demystifying the recent research. In particular, we will discuss methods such as Neral Radiance Fields (NeRF) and Neural Implicit Surfaces for multi-view 3D modeling. We will describe modifications and enhancements needed to adapt these methods from ground level imagery to aerial and satellite imagery. We will also review recent work in using AI to estimate height maps from a single overhead image.

Learning Outcomes: Attendees will:

  • Gain a better understanding of the current ways artificial intelligence (AI) and machine learning (ML) are revolutionizing the field of 3D reconstruction from remote sensing imagery. The focus is on demystifying these latest advances and making them more accessible to a broader audience. 
  • Be able to understand how these latest methods work at a very high level and gain some understanding about the advantages and disadvantages of different methods. 

While this is not a hands-on tutorial, we will provide references to open-source software packages that implement the methods discussed so that attendees can explore these tools in their own time.

Prerequisites: There are no hard prerequisites for this course. Attendees may benefit from having some prior knowledge about photogrammetry, multi-view geometry, or neural networks, but we will aim to cover this background material at a high level to make the course accessible to anyone.

Type: Training


Notes

Create or Log in to My Show Planner to add notes.


Speakers


Tracks