Make-A-Video

GPTKB entity

Statements (52)
Predicate Object
gptkbp:instanceOf text-to-video model
gptkbp:announced 2022-09-29
gptkbp:author Y. Singer, A. Polyak, T. Wolf, et al.
gptkbp:canAnimate static images
gptkbp:citation gptkb:Make-A-Video:_Text-to-Video_Generation_without_Text-Video_Data
2022
gptkbp:developedBy gptkb:Meta_AI
gptkbp:generation short video clips
animated scenes
realistic videos
stylized videos
videos at 64x64, 128x128, 256x256 resolutions
videos from image and text
videos from text
videos up to 5 seconds
videos with 16 frames
videos with different actions
videos with different artistic styles
videos with different backgrounds
videos with different camera motions
videos with different color palettes
videos with different compositions
videos with different environments
videos with different lighting conditions
videos with different objects
videos with different perspectives
videos with different subjects
videos with various styles
gptkbp:hasDemo yes
https://www.w3.org/2000/01/rdf-schema#label Make-A-Video
gptkbp:input text prompt
gptkbp:inputModality gptkb:text
gptkbp:language English
gptkbp:openSource no
gptkbp:organization gptkb:Meta_Platforms
gptkbp:output gptkb:DVD
gptkbp:outputModality gptkb:DVD
gptkbp:relatedTo gptkb:Imagen_Video
gptkb:Phenaki
gptkb:DALL-E
text-to-image models
gptkbp:researchPaper https://makeavideo.studio/static/paper.pdf
gptkbp:status research prototype
gptkbp:uses transformer architecture
diffusion model
latent diffusion
autoencoding
large-scale video datasets
pre-trained text-to-image models
gptkbp:website https://makeavideo.studio/
gptkbp:bfsParent gptkb:Meta_AI
gptkbp:bfsLayer 5