Google adds image-to-video generation to Veo 3
Google is continuing to advance its AI-powered video tools, announcing the rollout of an image-to-video generation feature in its Gemini app. This new capability is powered by Veo 3, Google’s latest AI video generation model.
The image-to-video feature allows users to turn still images into dynamic video clips directly through the Gemini app. By selecting the “Videos” option in the prompt box and uploading an image, users can generate short clips. There’s also an option to enhance these clips with sound by simply describing the desired audio in the text prompt. Once created, videos can be downloaded or shared.
This feature was initially introduced in May within Google's AI video tool called Flow, unveiled during the company's I/O developer conference. Since then, Google has gradually expanded access to Veo 3-powered video generation. As of last week, users in more than 150 countries can try out the feature, although access is currently limited to subscribers of the Google AI Ultra and AI Pro plans. These users are allowed to create up to three videos per day, with unused creations not carrying over to the next day.
Since Veo 3's launch seven weeks ago, users across Gemini and Flow have generated over 40 million videos, according to Google. To promote responsible use of its AI-generated content, Google applies both visible and invisible watermarks to all videos produced with Veo 3. Each clip is marked with a visible “Veo” logo and an invisible SynthID watermark, a technology Google developed to help identify AI-generated media.
With this update, Google continues its push to compete in the rapidly growing field of AI-driven creative tools, alongside other tech giants investing in generative media technologies.
