Amazon launches Titan Image Generator V2 with Advanced Image Conditioning for edits

Amazon has introduced an upgraded version of its AI image generator, Amazon Titan Image Generator V2, featuring advanced image conditioning capabilities, enabling users to detect and segment specific parts of an image for targeted edits and suggestions. This functionality is particularly useful for marketing and advertising campaigns, allowing businesses to create multiple versions or evolutions of an image for evaluation.

The upgraded model supports fine-tuning with assets such as company logos or specific products. This ensures that generated images align closely with brand guidelines. Users can provide a reference image along with a text prompt, giving them greater control over the final result by generating outputs that follow the specific layout or structure of the reference image.

Key features of the model include automatic background removal for images with multiple objects and a subject consistency tool that preserves specific subjects across different image variations. Currently available on Amazon Bedrock in the US East (N. Virginia) and US West (Oregon) AWS Regions, there are plans for future expansion to other regions.