Google has developed a new AI model, Lumiere, for video generation and editing that utilises the Space-Time-U-Net (STUNet) diffusion model.
Lumiere can create videos in one process, instead of stitching smaller frames together, by approximating where objects within a frame will move. It generates 80 frames compared to the 25 generated by competitors like Stable Video Diffusion.
Lumiere will also allow for image-to-video generation, stylised generation, cinemagraphs and inpainting. While Lumiere is not yet available for testing, Google’s slight presence in the text-to-video category will likely change with its advanced AI model.
The company has acknowledged the risk of misuse, and emphasised the importance of detecting biases and malicious use cases to ensure a safe and fair use.
@wikitechleaks
#LumiereAi #ai #artificialintelligence #technews #techupdates #latesttechnews #technology #trends #trendingnews #wikitechleaks #latestupdates #technologyupdates #technologytrends #aivideocreator #LumiereAiVideo #GooglesLumiere #AiVideo #TexttoVideo #PhototoVideo #cinemagraphs #advancedAI #GenerativeAI #STUNet #diffusionmodel #stablevideodiffusion #google #aipower #openai #BreakingNews
GDQ事务所原创资源，gdqun.com
