2 Sources
[1]
Google Gemini learns a new trick using reference images for Veo 3.1 video creations
While it's been a few years since the big boom of AI, the tools still keep coming. And each step of the way we're left speechless with what's being accomplished, only to ponder how things might look in just a few years time. Despite these major advances, it still feels like we're in the early
[2]
Gemini app gets a new way to guide photo-to-video generations
Google is updating the Gemini app today with a better way to use photo-to-video generation by adding visual ingredients. You can now upload up to three reference images to guide Veo in the Gemini app. These visual "ingredients" -- like in the Flow video tool -- can be of characters, objects,
Share
Copy Link
Google rolls out a new update to its Veo 3.1 video generation tool within Gemini, allowing users to upload up to three reference images as visual 'ingredients' to guide AI video creation. The feature aims to reduce reliance on complex text prompts while enabling more personalized and accurate video outputs.
Google has announced a significant update to its Veo 3.1 video generation tool within the Gemini app, introducing the ability to use reference images as visual guides for AI-powered video creation. The new feature, which Google calls visual "ingredients," allows users to upload up to three reference images to better direct the AI's video generation process
1
2
.The reference image functionality enables users to incorporate specific characters, objects, styles, and scenes into their video prompts by providing visual examples rather than relying solely on text descriptions. This approach mirrors the capabilities found in Google's Flow video tool and represents a shift toward more intuitive content creation methods
2
.According to Google's demonstration, users can upload images of characters and have them placed into specified scenes, where they will act according to the accompanying text prompt. This integration of visual and textual guidance aims to produce more accurate results on the first attempt, reducing the need for multiple iterations
1
.
Source: 9to5Google
Google positions this update as a solution to the complexity of crafting detailed text prompts for video generation. By allowing users to provide visual references, the company hopes to make the creative process more accessible and efficient. The feature also enables creators to incorporate original content into their videos while letting Veo handle the technical aspects of video generation
1
.The update adds a personal dimension to AI video creation, as users can now include their own imagery and have the AI system build upon these visual foundations. This capability extends beyond simple entertainment applications, offering new tools for creative expression and content development
1
.Related Stories
The reference image feature is rolling out starting today across both mobile and desktop platforms, with full availability expected next week. Access to this functionality requires a subscription to Google AI Plus, Pro, or Ultra plans, though the basic Gemini account provides access to standard video generation tools
1
2
.Google offers complimentary access to premium AI features for new Pixel phone owners and students, expanding the potential user base for these advanced capabilities. The company has also updated the Gemini Tools menu on Android and iOS to clearly specify when users are utilizing the Veo 3.1 model for video generation
1
2
.This update represents part of Google's broader strategy to establish Veo 3 as the leading platform for AI-generated video content. The company recently demonstrated confidence in its technology by creating and releasing an advertisement produced entirely using its own AI tools, showcasing results that closely resemble professionally produced content
1
.Summarized by
Navi
[1]
1
Science and Research

2
Policy and Regulation

3
Technology