BHAVIK CHANDNA retweetledi

Learned priors are useful in improving the discrete, noisy 3D predictions from DUSt3R/MASt3R. This paper has been added to the 🚀awesome-dust3r🚀 list, check out for more details at: github.com/ruili3/awesome…
Zhenjun Zhao@zhenjun_zhao
Spurfies: Sparse Surface Reconstruction using Local Geometry Priors Kevin Raj, Christopher Wewer, Raza Yunus, @EddyIlg, @janericlenssen tl;dr: neural point representation disentangles geometry & appearance; pre-train geometry prior on synthetic dataset arxiv.org/pdf/2408.16544
English





















