Google has introduced an innovative feature within its Gemini AI suite, enabling users to convert photos into video clips. This new capability, known as photo-to-video, utilizes Google’s Veo 3 video model to transform still images into eight-second videos, complete with AI-generated audio that includes background sounds, environmental effects, and spoken dialogue.
Currently, the Gemini video feature is accessible to Google AI Ultra and Pro subscribers in selected regions. The rollout begins today on web platforms and will extend to mobile devices throughout this week.
To utilize this feature, Gemini users can navigate to the “tools” menu in the prompt bar, select the “video” option, and upload a photo, along with a text description outlining how the image should animate. Users can also specify audio elements for dialogue, sound effects, and background noise, all synchronized seamlessly with the visuals. The resulting videos are delivered in MP4 format with a resolution of 720p in a 16:9 landscape layout.
Google encourages creativity by allowing users to animate ordinary objects, enliven their drawings or paintings, or introduce motion in nature scenes. All videos generated through this feature will bear a visible watermark indicating they are AI-produced, along with an invisible SynthID digital watermark.
While a comparable feature was launched with Flow, Google’s generative AI filmmaking tool earlier this year, the new capability in Gemini allows users to animate their photographs directly without the need for an external application. Additionally, the Flow tool is set to become available in 75 more countries today, coinciding with the expansion of Gemini’s video feature.
