Free AI Image Generator

00:00:01 - 00:01:49 Challenges of consistent AI character creation

00:00:01 - 00:01:49 The video discusses the challenge of maintaining a consistent character appearance across multiple AI-generated images, highlighting common issues like varying facial features, hairstyles, and colors. Despite attempts using fixed seed numbers in Leonardo AI, consistency remains elusive, which is frustrating for storytelling or brand building. The presenter promises a free method to create an AI character that looks identical across different poses, scenes, and settings.

00:01:22 - 00:02:23 Crafting a detailed character prompt with ChatGPT

00:01:22 - 00:02:23 The segment emphasizes the importance of starting with a strong foundation when creating a consistent character. It advises crafting a detailed character reference that goes beyond simple traits, using ChatGPT to generate a comprehensive and realistic image prompt including face, skin tone, hair texture, outfit, personality vibes, and style.

00:01:52 - 00:03:43 Using Google Image FX for AI image generation

00:01:52 - 00:03:43 The narrator explains the importance of providing a detailed description to help the AI accurately generate character images. They introduce Google Image FX as a free and effective tool for AI image generation, mentioning Google's VO3 for text-to-video creation. The process involves pasting a full prompt from ChatGPT into Image FX, enhancing it with style keywords like hyper-realistic and cinematic lighting to ensure a consistent aesthetic, then generating the character image.

00:03:06 - 00:04:48 Locking the seed for consistent character visuals

00:03:06 - 00:04:14 The speaker explains how to lock the seed number next to the image preview to preserve the character's unique visual features during regeneration. Locking the seed ensures that the face, features, and overall vibe remain consistent with each new image generated, preventing unwanted changes even if the prompt stays the same.

00:03:41 - 00:04:48 After locking the seed, the speaker demonstrates generating images that maintain the same character's face and vibe. The next step involves placing this consistent character, named Looda, into different scenes. They proceed to use ChatGPT to create a scene description of Looda going for a morning walk on the street, preparing to integrate her into new environments.

00:04:20 - 00:05:30 Placing character in different scenes consistently

00:04:20 - 00:05:30 The segment explains how to modify a character's action in Image FX while maintaining the original character description. It demonstrates generating images of the character Loa in different scenarios—walking and sitting after a gym session—highlighting the consistency in appearance despite changes in action. It then suggests using ChatGPT to create a story about Loa, a young girl from a poor family.

00:04:57 - 00:06:11 Building visual stories with consistent AI character

00:04:57 - 00:06:11 The speaker explains how to use ChatGPT to create a rich, visual story by first generating a 200-word story prompt. Then, the story is broken into key moments suitable for visuals. Each scene description is combined with the original prompt to generate consistent images in image FX, maintaining the same character appearance throughout. This method is ideal for creating comics, video storytelling, or Instagram carousels.

00:05:36 - 00:07:15 Summary: steps for consistent AI character creation

00:05:36 - 00:07:15 The tutorial explains how to create a consistent AI character for storytelling without paid tools. It recommends starting with a detailed character prompt using ChatGPT for rich descriptions and adding style keywords for consistent, professional-looking images. Locking the seed number in Image FX ensures the character's face remains the same across images. The core character description should remain unchanged while new scenes are added one at a time. This method allows you to generate consistent AI characters across various settings, poses, and scenes without advanced tools or extra cost. Viewers are encouraged to try the method and share their results.

00:06:42 - 00:06:58 Closing and call to action for viewers

00:06:42 - 00:06:58 The video concludes by encouraging viewers to like and subscribe for more AI tutorials, thanking them for watching and wishing them happy creating.