2 papers
cs.CV2026
LucidLyra: Feed-Forward Generative 3D Scene Reconstruction and Joint Understanding via Multi-Task Distillation
Oliver Hahn, Rodrigo Marcuzzi, Sangeetha Grama Srinivasan +7
We present LucidLyra, a framework that makes generatively reconstructed 3D scenes directly queryable for semantics and instances in a feed-forward manner from video latents. While…
cs.CV2026
Déjà View: Looping Transformers for Multi-View 3D Reconstruction
Alessandro Burzio, Tobias Fischer, Sven Elflein +9
Recent feed-forward 3D reconstruction transformers have scaled to over a billion parameters, following the broader trend of increasing model capacity in computer vision. Yet emergi…