The model was designed to tackle a significant challenge in video generation synthesis, which is creating 'realistic, diverse, and coherent motion,' according to the paper.
Lumiere can generate video from different inputs, including text-to-video and image-to-video, as well as stylized generation using a single reference image.
#google-research #lumiere #ai-video-generation #text-to-video-diffusion-model #space-time-u-net-architecture
[
add
]
[
|
|
...
]