View Proposal
-
Proposer
-
Phil Bartie
-
Title
-
Rapid 3D mapping from single camera
-
Goal
-
To produce a 3D point cloud of a scene from a single camera
-
Description
- Structure from Motion require many photos from different angles, but with the AI monocular depth libraries there are new ways to build detailled scene maps from fewer images (eg Yolov26-Depth; COLMAP; Depth Anything v3). This project will develop and evalutate the current best libraries to produce a data pipeline which takes a single video and exports a 3D model / point cloud.
- Resources
-
Video camera
Python + GPU (eg Google Colab)
-
Background
-
-
Url
-
-
Difficulty Level
-
Moderate
-
Ethical Approval
-
None
-
Number Of Students
-
0
-
Supervisor
-
Phil Bartie
-
Keywords
-
-
Degrees
-
Master of Science in Artificial Intelligence
Master of Science in Data Science
Master of Science in Robotics