Nvidia introduces Axolotl3D for 3D reconstruction
Nvidia says the model combines images, camera data and partial geometry to reconstruct missing regions of objects while preserving the parts already observed.
TLDR
Nvidia introduced Axolotl3D at ECCV 2026, describing it as a 3D generation model that accounts for hidden parts of objects. The company says it uses images, camera data and partial geometry to reconstruct missing regions while preserving observed ones, and claims state-of-the-art results across single- and multi-view settings.
Combined views
26.2K
1 Source, first seen ago