Reproduce “Video Depth Estimation by Fusing Flow-to-Depth Proposals” 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
Jiaxin Xie, Chenyang Lei, Zhuwen Li, Li Erran Li, Qifeng Chen
The authors estimate dense depth from monocular video by triangulating depth proposals from optical-flow correspondences and camera poses, then fusing proposals, confidence maps, and image content with a supervised network. On KITTI’s Eigen split, their supervised-plus-video model reports 0.081 abs rel versus 0.102 for DORN. On KITTI-to-Waymo, it reports 0.150 versus NeuralRGBD’s 0.177, a system-level result on this cross-dataset test.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.