Video models ❤️ 3D? Room 607
Yilun Du (Harvard) at #CVPR2026 http://geofreenvs.github.io
8:09 AM · Jun 4, 2026 · 2.5K Views
Harvard's Yilun Du presents Unified World Models for reconstructing complete 3D scenes from partial text or video inputs
The project powers geometry-free novel view synthesis.
--0--
Sentiment
Sentiment building, check back later.
Cluster Engagement
-
Views
-
Comments
-
Reposts
-
Bookmarks
Expand data
Posts from X
Most Activity
Most Activity
VIEWS555LIKES2
Kosta Derpanis (sabbatical in Zurich)@CSProfKGD
@taiyasaki @du_yilun
Andrea Tagliasacchi @CVPR@taiyasaki
Video models ❤️ 3D? Room 607
Yilun Du (Harvard) at #CVPR2026 http://geofreenvs.github.io
8dViews 555Likes 2Bookmarks 0