Timezone: »

 
Poster
Text-To-4D Dynamic Scene Generation
Uriel Singer · Shelly Sheynin · Adam Polyak · Oron Ashual · Iurii Makarov · Filippos Kokkinos · Naman Goyal · Andrea Vedaldi · Devi Parikh · Justin Johnson · Yaniv Taigman

Wed Jul 26 05:00 PM -- 06:30 PM (PDT) @ Exhibit Hall 1 #301
in Poster Session 4 »
Event URL: https://make-a-video3d.github.io/ »

We 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. The dynamic video output generated from the provided text can be viewed from any camera location and angle, and can be composited into any 3D environment. MAV3D does not require any 3D or 4D data and the T2V model is trained only on Text-Image pairs and unlabeled videos. We demonstrate the effectiveness of our approach using comprehensive quantitative and qualitative experiments and show an improvement over previously established internal baselines. To the best of our knowledge, our method is the first to generate 3D dynamic scenes given a text description. Generated samples can be viewed at make-a-video3d.github.io

[ PDF]  [ Poster] 

Author Information

Uriel Singer (Meta AI)
Shelly Sheynin (Meta)
Adam Polyak (Meta)
Oron Ashual (Meta)
Iurii Makarov (Facebook)
Filippos Kokkinos (Meta AI)
Naman Goyal (Facebook)
Andrea Vedaldi (University of Oxford)
Devi Parikh (Georgia Tech & Facebook AI Research)
Justin Johnson (University of Michigan)
Yaniv Taigman (Meta)

More from the Same Authors