Heaptalk, Jakarta — Amazon issued the latest version of its AI image-generating model, Titan Image Generator v2, available on the Bedrock generative AI platform (08/06).
This upgrade claimed to streamline workflows, boost productivity, and bring creative visions to life. In this latest version of Titan, users can guide image creation using reference images, edit existing visuals, remove backgrounds, generate image variations, and securely customize the model to maintain brand style and subject consistency.
Titan v2 supports two types of image conditioning: canny edge and segmentation. The canny edge algorithm extracts the prominent edges within the reference image, creating a map that the image generator can use to guide the generation process. Segmentation provides a more granular level of control. By supplying the reference image, users can define specific areas or objects within the image and instruct Titan to generate content that aligns with those defined regions.
Furthermore, this model is able to instantly remove the image background with a single step. “Amazon Titan Image Generator v2 can intelligently detect and segment multiple foreground objects, ensuring that even complex scenes with overlapping elements are cleanly isolated,” Channy Yun, Principal Developer Advocate for AWS cloud, stated on the company’s official blog (08/06).
The example shows an image of an iguana sitting on a tree in a forest. The model identified the iguana as the main object and removed the forest background, replacing it with a transparent background. This lets the iguana stand out clearly without the distracting forest around it.
