Posted on by

Google is testing picture-to-video mode with scene controls in Veo

Google is testing picture-to-video mode with scene controls in Veo

Google’s VideoFX and Veo are making waves in the AI-generated video landscape, with recent discoveries pointing to a separate image-to-video mode that includes scene and motion controls. This development, uncovered by TestingCatalog, suggests that Google is pushing the boundaries of its Veo generative video model to offer more versatile and controlled video creation capabilities.

The new image-to-video mode allows users to provide a prompt to generate images using Imagen3, Google DeepMind’s highest quality image generation model. This approach is reminiscent of the previous “Storyboard” implementation, which was removed but now seems to be making a comeback with expanded functionality. The scene control function can generate movements based on the original object or scene, providing a more dynamic and realistic video creation experience.

Quick controls are also available with various drop-down options, indicating a more user-friendly interface. Although the Veo generation is not yet fully functional, the changes suggest that it is in Google’s trusted testing phase, suggesting a possible release in the near future.

Google’s commitment to developing responsible AI tools in collaboration with creators and artists is evident in updates to VideoFX, ImageFX, and MusicFX. These tools are designed to support creatives on the storytelling journey and provide more control and flexibility in the generative creation process. The introduction of My Library, which allows users to save, review and remix content, further improves the creative workflow.

Competition in the video generation space is heating up, with OpenAI’s Sora and LumaLabs’ Dream Machine also vying for attention. However, Google’s comprehensive approach to video creation, which combines video generation with a timeline feature and AI audio capabilities, could give the company an advantage in the market.

As VideoFX and Veo continue to evolve, it will be interesting to see how these tools shape the future of video creation and storytelling. As Google focuses on responsible AI development and collaboration with creators, the opportunities for innovative and high-quality video content are enormous.

Key features of the new picture-to-video mode include:

  • Image to video mode: Allows users to specify a prompt to generate images using Imagen3.
  • Scene control: Creates motion based on the original object or scene.
  • Quick checks: Provides various drop-down options for a more user-friendly interface.
  • Veo generation: Running in Google’s trusted testing phase, indicating a possible release soon.

These developments underscore Google’s commitment to advancing AI-generated video capabilities and giving creators more tools to bring their ideas to life.