EpiDiffVO uses sparse epipolar matching, diffusion-based refinement, and graph neural network subset selection to reduce correspondence redundancy while maintaining robust relative pose estimation on TartanAir and KITTI datasets.
Stereoglue: Robust estimation with single-point solvers,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
EpiDiffVO: Geometry-Aware Epipolar Diffusion for Robust Visual Odometry
EpiDiffVO uses sparse epipolar matching, diffusion-based refinement, and graph neural network subset selection to reduce correspondence redundancy while maintaining robust relative pose estimation on TartanAir and KITTI datasets.