First teased back in May, Google’s latest generative AI video model officially launched in a private preview. Veo is noteworthy as it was announced around the same time as OpenAI’s Sora (which has perhaps turned heads more than any other AI video model to date), yet has now beat Sora to market.And with OpenAI still not announcing any release date (or updated plans for that matter) on when Sora might come out, those interested in exploring AI for video are left to utilize what they currently have in front of them.And with Google Veo, we appear to have a quite sophisticated AI video model that is capable of generating what the company dubs as “high-quality” videos at up to 1080p resolution across a range of different visual and cinematic styles. Let’s take a look at this new text or image prompt based video model and its generative AI possibilities.Google Veo AI Video ModelDeveloped by Google’s DeepMind initiative, Veo is here to generate high-quality, high-definition videos based on text or image prompts. The AI model is capable of producing a range of cinematic and visual styles at pretty exceptional speeds. Veo features an advanced understanding of natural language and visual semantics which should set it apart from some of its main AI video competition and has really been tailored to closely align with the nature of its prompts. Google’s Veo on Vertex AI further creates footage that’s consistent and coherent so creators can expect people, animals, and objects to move realistically...
Published By: NoFilmSchool - Wednesday, 4 December, 2024