Lab: Personality Drawing
- Corrina Crazie Espinosa
- Jan 21
- 5 min read
Updated: Jan 29

Now that you have taken the personality quiz and you have an idea who your digital doppelganger is, and a loose idea of what they look like, it's time to generate some content!
In this lab, we will create various content to design a character in 't-pose' using Photoshop. This character will be used to prepare a puppet for Adobe Character Animator, a 2D animation tool
We will also explore using AI to convert your character into a 3D model. This project focuses more on video and image generation, so we won't delve deeply into looping 3D animations (like Mixamo). We'll cover that in the next project, which involves creating a 3D world.
I am using the famous Great Depression-era photograph by Dorthea Lange called Migrant Mother as the base image for my personality drawing. You can use whatever you like including photos, IRL drawings, paintings, digital drawings (procreate, illustrator), generated images, 3D models, etc.
AI avatar generators
https://www.lightxeditor.com/ - Generate quick stylized avatars using your own image as a reference. you get 5 free credits to start and can purchase 500 credits for $4.99 if you like it. It also has some quick video meme generators that are worth looking at.
Composer - Artbreeder Generate a lot of poses from your base image, fine tune with prompts, combine multiple images. Some results were unexpected, like this random snake?! I don't know why the robots do what the robots do sometimes.
You can get a wide variety of images very fast for free and use prompts describing T-pose to get in that ballpark. Beware, it can get a little weird, my character slowly started to turn masculine, so I added the word "feminine" to the prompt, and I ended up getting some bare breasts and nudity. Also remember, people are never static. They change rapidly, so you could follow the path of transformation and let it contribute to the overall story that is your alter ego. **This example could be a poverty stricken mother who took her rage out in the gym and became a world class body builder... until she slipped down the dark path of steroid usage which destroyed her career but led to her ultimate self actualization as He began to live as his genuine self, a male businessman.
https://playground.com/ - Quickly edit your refrence image with many different free styles (paid styles too). Use text to make specific edits, like mood, or replacing objects. Be aware you do get a limited amount of generations-- about a dozen, before you get cut off and have to pay, they have a variety of packages, including an $8 unlimited day pass.
https://app.runwayml.com/ By far one of the best video AI tools around, lots of tools and options, including lip-sync. You can add typed text and select a voice, or you can upload your own audio. This can be an expensive tool, but you can also get a lot of freebies out of it. Worth trying out for sure.
Luma Dream Machine: New Freedoms of Imagination Another powerful AI video tool that let's you upload images to convert to videos along with a text description. Slightly different style from Runway ML, I find this tool to be a little more surreal and abstract, though you can also create very realistic videos. It's a little more affordable and also has some free play. Just like with Runway ML you can upload a starting and ending keyframe, type in a prompt to get your results.
Stable Video Diffusion 1.1 - a Hugging Face Space by multimodalart Hugging Face is an open source platform with tons of AI generators to play with. This one is specifically turning image into video. Mostly free, with limitations on GPU usage.
Here are some other cool hugging face tools to try:
AI Video Generator Another really simple video generator that let's you feed in your own base image and write a text prompt.
https://magic.remix.ai/ A magic camera to add your own likeness to your alter ego, with a text prompts to control the narrative.
Hailuo AI: Transform Idea to Visual with AI A Chinese based video generator that let's you use base image + prompts and starts you out with some free credits.
MyHeritage Deep Nostalgia™, deep learning technology to animate the faces in still family photos - MyHeritage Specifically for animating portraits to look alive.
sudoAI - Instantly turn images into 3D models! Not perfectly accurate, but an excellent start that you can then edit in Blender or use as is if you like the serendipitous quality of the models. Paid, but you get free credits to try it out.
Community - Meshy Like Sudo, but BETTER, I think it has better options and better results. Paid, but let's you try for free.
3D Model Generator another one! Similar, might not geneate as high of quality as meshy, but it let's you try for free. MILK those free credits~
https://www.flexclip.com/ Tons of AI tools to try, including a 2D image to 3D video generator... It's not fantastic, but the tech is still so new it will be interesting to watch how it develops.
Daz 3D - 3D Models and 3D Software | Daz 3D A free app that lets you design 3D characters in an environment similar to Blender, but much simpler and specific to character design. It's fairly easy to get started and there are a lot of tutorials to pursue if this is interesting to you. This is NOT an AI based software. It does limit creative freedom a little bit and is a lot more "template" based than Blender but can get you to your end goal much quicker and easier.

This is a SHORT list of tools available to you. I tried to pick the ones that are most relevant to the project, that had at least some freebies, and didn't require a credit card to try. The nature of this medium is that it is very new and changing very quickly. These sites can be up or down, and the parameters can change very rapidly, so this list could be worthless in a year from now.
Next, let's take the materials we have gathered and make a final form in T-pose that you will use for the lab next week when we learn to prep our puppets for Adobe Character Animator. I will record the tutorial and post it for your reference.
Homework: For the lab next week, you should have a final version of your character's full body in T-Pose completed.
Kommentare