Ruojin Cai (@ruojin8) 's Twitter Profile
Ruojin Cai

@ruojin8

PhD student at Cornell CS

ID: 1387862520757723138

linkhttps://www.cs.cornell.edu/~ruojin/ calendar_today29-04-2021 20:13:14

18 Tweet

282 Followers

135 Following

Ruojin Cai (@ruojin8) 's Twitter Profile Photo

Can we estimate relative rotations between *non-overlapping* images? Yes! Our new #CVPR2021 paper uses 4D correlation volumes to reason about implicit cues that reveal their geometric relationship. With Bharath Hariharan Noah Snavely @elorhadar. Project page: ruojincai.github.io/ExtremeRotatio…

Can we estimate relative rotations between *non-overlapping* images?
Yes!
Our new #CVPR2021 paper uses 4D correlation volumes to reason about implicit cues that reveal their geometric relationship. With <a href="/BharathHarihar3/">Bharath Hariharan</a> <a href="/Jimantha/">Noah Snavely</a> @elorhadar.
Project page: ruojincai.github.io/ExtremeRotatio…
Gene Chou (@gene_ch0u) 's Twitter Profile Photo

Introducing MegaScenes—a scene-level dataset containing 100K SfM reconstructions and 2M images with open content licenses. We validate its effectiveness in training large-scale, generalizable models on the task of novel view synthesis. 1/N project page: megascenes.github.io

Dmytro Mishkin 🇺🇦 (@ducha_aiki) 's Twitter Profile Photo

Extreme Rotation Estimation in the Wild Hana Bezalel, Dotan Ankri, Ruojin Cai Hadar Averbuch-Elor tl;dr: MegaDepth/Scenes subset with small/large/no overlap image pairs, the task is R prediction arxiv.org/abs/2411.07096

Extreme Rotation Estimation in the Wild
Hana Bezalel, Dotan Ankri, <a href="/ruojin8/">Ruojin Cai</a> <a href="/ElorHadar/">Hadar Averbuch-Elor</a> 

tl;dr: MegaDepth/Scenes subset with small/large/no overlap image pairs, the task is R prediction
arxiv.org/abs/2411.07096
Dmytro Mishkin 🇺🇦 (@ducha_aiki) 's Twitter Profile Photo

Doppelgangers++: Improved Visual Disambiguation with Geometric 3D Features Yuanbo Xiangli, Ruojin Cai, Hanyu Chen, Jeffrey Byrne, Noah Snavely tl;dr: new dataset (55K pairs) + Mast3r == PROFIT arxiv.org/abs/2412.05826

Doppelgangers++: Improved Visual Disambiguation with Geometric 3D Features 

Yuanbo Xiangli, Ruojin Cai, Hanyu Chen, Jeffrey Byrne, <a href="/Jimantha/">Noah Snavely</a> 
tl;dr: new dataset (55K pairs) + Mast3r == PROFIT
arxiv.org/abs/2412.05826
Yuanbo Xiangli (@ambie_kk) 's Twitter Profile Photo

Introducing Doppelgangers++! 🚀 An enhanced pairwise image classifier that tackles visual aliasing (doppelgangers) to improve 3D reconstruction accuracy across diverse, real-world scenes. 🌍✨ 🔗Project page: bit.ly/3VAPMJc. Code is also available.