Reproduce “Rain Beyond Streaks: Cascaded Realistic Rain and Wet Surface Synthesis via Scene Priors” 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
Melissa Tijink, Kirill Tolstobrov, Shunxin Wang, Estefania Talavera, Duc V. Le, Luuk Spreeuwers, Shaodi You, Nicola Strisciuglio
CascadeRS synthesizes controllable rain scenes by using pretrained depth and semantic segmentation priors to add reflections, puddles, fog, and streaks to clear images. In evaluations, it received 2015 human ranking points versus TPSeNCE’s 1718, and improved real-rain ACDC FWIoU for both tested segmentation models. However, these results do not establish universal realism or consistent detection gains.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.