A learned fusion of frames and event-camera data removes synthetic snow in DSEC-Snow with a 3 dB PSNR gain, but it does not beat image-only baselines on the paper's real snowfall dataset.
Deep dense multi-scale network for snow removal using semantic and geometric priors,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
REJECT 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Event-Based De-Snowing for Autonomous Driving
A learned fusion of frames and event-camera data removes synthetic snow in DSEC-Snow with a 3 dB PSNR gain, but it does not beat image-only baselines on the paper's real snowfall dataset.