Decomposing scene motion into per-point 6DoF motion maps from optical flow and depth enables a neural network to estimate camera motion more accurately than stacking raw inputs.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Scene Motion Decomposition for Learnable Visual Odometry
Decomposing scene motion into per-point 6DoF motion maps from optical flow and depth enables a neural network to estimate camera motion more accurately than stacking raw inputs.