- Google has announced a major update to “Photo-to-video,” the feature that generates videos from photos, currently rolling out for the Android and iOS versions of the “Google Photos” app.
- A text prompt feature has been added.
- Users can now provide more specific instructions through “custom prompts” to specify movements or styles, and the ability to edit suggested prompts allows for even more detailed input.
On Tuesday, January 27, 2026, Google announced a major update to “Photo-to-video,” a feature that generates videos from photos. This feature originally began its US rollout for the Android and iOS versions of the “Google Photos” online photo library service on Wednesday, July 23, 2025.
“Photo-to-video” is nearly identical to the feature of the same name launched on Thursday, July 10, 2025, as a new function for the Google AI “Gemini.” It uses AI to generate a 6-second video based on a photo. Specifically, users select a photo and simply choose one of two suggested prompts.
Videos created using “Photo-to-video” are embedded with a “SynthID” digital watermark to indicate that they were created using generative AI. Additionally, in September 2025, the video generation AI model used for “Photo-to-video” was upgraded from the original “Veo 2” to “Veo 3.”
In this major update, a text prompt feature has been added to “Photo-to-video.” While the feature previously required users to choose from two suggested prompts, it now allows for more detailed instructions. This includes “custom prompts” where users can specify specific movements or styles, as well as the ability to re-edit suggested prompts to provide more concrete directions.
That said, “Photo-to-video” is still currently limited to the US market. The rollout has not yet been expanded to other regions.
“Google Photos” App Links
Source：Google
コメントを残す