Text-To-4D
Meta present MAV3D (Make-A-Video3D), a method for generating three-dimensional dynamic scenes from text descriptions. Our approach uses a 4D dynamic Neural Radiance Field (NeRF), which is optimized for scene appearance, density, and motion consistency by querying a Text-to-Video (T2V) diffusion-based model.
Website: https://make-a-video3d.github.io
In category: 4D Generation
Page views: 620
(Based on 0 ratings)
0 vote
0 vote
0 vote
0 vote
0 vote
Member's review
No ratings has recorded yet. Be the first to rate it.