Reproduce “DAPEVO: Deep Adaptive Patch Frame-Event Visual Odometry” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Luca Gandolfi, Simone Nascivera, Roberto Pellerito, Rong Zou, Chiara Plizzari, Davide Scaramuzza
DAPEVO separately matches RGB and event measurements, then adaptively weights their evidence for pose and depth estimation. When retaining one in six UZH-FPV RGB frames, mean ATE rises from 1.00 to 1.36 m versus 3.7× and 3.1× increases for DPVO and RAMP-VO. On degraded synthetic TartanEvent it reaches 0.60 m. This supports sparse-RGB and synthetic-degradation tolerance, not real-world low-light robustness.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.