How to Create Consistent Cartoon Characters with Next Diffusion

In the world of digital art, creating fun cartoon characters requires both creativity and the right tools. Next Diffusion is one such tool. It's an AI image generator that transforms how we create characters. This introduction marks the beginning of our exploration into cartoon character creation. We'll see how Next Diffusion makes it easy and precise to bring characters to life. With Next Diffusion, users can select the perfect pose and swap faces effortlessly. Let's discover how to use Next Diffusion to craft fantastic cartoon characters!

After logging in and entering the dashboard, we can begin our quest to create cohesive cartoon characters. The initial phase involves choosing a "Cartoon" checkpoint. Head over to the "Checkpoint" subtab, offering numerous selections. Specifically, for our objective, we'll pick the "ReVAnimated" checkpoint, which serves as the foundational reference for the image style.

Next, let's fine-tune some base settings located on the left side of the dashboard. Since we've chosen a cartoon checkpoint, it's best to use the "Euler A" sampler for smoother results. For sampling steps, a value of 30 strikes a balance between detail and render time, though remember, higher sampling steps mean more detailed images but longer rendering times.

• Sampler: Euler a
• Sampling Steps: 30
• Batch Size: 2
• File Format: jpg
• Aspect Ratio: 1:1

Adjust the batch size to 2 and select jpg as the file format. The default aspect ratio (1:1) will suffice for our needs. With these adjustments in place, we're primed to embark on our creative journey with Next Diffusion. Before clicking on generate, let's navigate to the "Pose" tab in the top navbar.

Next, proceed to the "Pose" subtab to choose the pose that aligns with your vision. While this step is optional, it enhances consistency in crafting cartoon characters. Here, you'll find various options such as laying, portrait, selfie, sitting, or standing poses. Just click on the image of your preferred pose. When a green checkmark appears, it signifies your chosen pose for the image generation. Should you wish to change or remove a selected pose, simply click on the image again to deselect it. For our project, we'll choose a sitting pose to ensure consistency in the final output.

Note: Fresh poses are added weekly to ensure your character stays dynamic with new poses regularly.

Following that, let's proceed to the "Face" upload section, where you have the option to upload a face for your cartoon character. Easily drag and drop your chosen image onto the canvas, then confirm the upload by clicking the corresponding button. Upon successful upload, a checkbox labeled "Use image as reference" will emerge beneath the canvas. If the selected face is not realistic and you're not utilizing a realistic checkpoint, it's advisable to enable this checkbox.

Note: The "Face" section is specifically designed for realistic face swaps. Therefore, when using a non-realistic face and a non-realistic checkpoint, it's crucial to enable the "Use image as reference" checkbox. Failure to do so will result in the tool attempting to create a realistic face swap from the cartoon image provided on the canvas, which may not yield accurate results. Enabling this checkbox ensures that the image is used as a reference, optimizing the generation process for non-realistic characters.

Next, let's craft a prompt describing the final image. In our scenario, the uploaded (face) image serves as a reference for the final output. As we've selected the "Use image as reference" checkbox in the "Face" tab, it's essential to note that the tool primarily relies on this image for guidance. However, if there are specific features you'd like to include in the final image, you can list them separated by commas within the "prompt" text field.

• Prompt: 30 year old woman,pink eyes, green hair, indoor, bed, pillows in background, gamer girl, tattoos , piercings, cat ears, looking in the camera, front view, sitting pose
• Negative Prompt: worst quality, low quality, lowres, monochrome, greyscale, bad anatomy, deformed, disfigured, mutation hands, mutation fingers, extra fingers, missing fingers, watermark

If all the settings are correct we can click on "Generate" and we can witness our cartoon character come to life.

After the generation process concludes, head to the "Generation Output" subtab to view your generated images. Click on the image to examine your image settings. You'll notice that the final output image incorporates the pose you selected and the reference face image.

Example 1

Example 2

Example 3

Crafting consistent cartoon characters has never been easier, thanks to Next Diffusion's AI image generation tool. By leveraging the intuitive interface and powerful features, users can bring their creative visions to life with remarkable precision. Whether you're an experienced artist or just starting, Next Diffusion offers the tools you need to unlock your creativity and produce stunning cartoon characters. With customizable settings, pose options, and face swapping capabilities, the possibilities are endless. Elevate your cartoon character creations today with Next Diffusion.