AFT-VO: Asynchronous Fusion Transformers for Multi-View Visual Odometry Estimation.

Nimet Kaygusuz Oscar Mendez Richard Bowden

Published in: CoRR (2022)

Keyphrases

multi view
visual odometry
single view
multiple views
d objects
depth map
ego motion
dynamic scenes
camera pose
three dimensional
camera calibration
multiple cameras
depth images
robust estimation
semi supervised
range images
d scene
multi camera
surface reconstruction
least squares
optical flow