Projects
Digital
Mixed-Media
- Material Perception: Analogue to 3D Model
- Misfit Dollz: The Exhibit (2025)
- Misfit Dollz: The Wonderverse (2026)
Misfit Dollz: The Wonderverse
Misfit Dollz: The Wonderverse is a multi-component mixed-media interactive experience that explores toys as culture signifiers and artifacts, as mediums of artivism, and their role in nostalgic and collaborative play practices. This page serves as a process-and-progress blog.
Materials & Mediums:
Digital: Unity, Tripo, Meshy, Polycam, Blender, Procreate, Canva, Final Cut
Physical/Analog: Paper, Scanner, Laptop, Mouse, Keyboard, Double-sided gloss paper, Sticker paper
Week 14: Installing & Finalizing the Wonderverse
While rebuilding the virtual space was time-sensitive, sourcing the materials and props for staging was as well. Now that I have established how I would lay out my environment, what my aesthetic inspirations are, and looked at the supplies I already have, such as tinsel, the toys used in the virtual environment, TV, and string lights. This made it easier to determine what I needed to get, which was as follows: x2 lower folding tables, a mouse and keyboard, 3 reading lights, a metallic tablecloth and skirt, pencil holders for my paper doll parts, and a rolling cart. After measuring the table and overall space, I determined I would need fold-up tables that were just shy of 3 feet, and I found great tables that also match the maximalist aesthetic I was aiming for. They were pink, so no table covering was needed! In case those tables didn’t fit, I also got one larger black picnic table that I would have perpendicular to the main display table. I also thought of ways I could further transform the space, like adding a rug or mat and more whimsical elements, so I went back to Jackpot Deals, among other dollar stores, to find supplies, as well as looked online. I found the following items to help with bring my aesthetic vision to fruition: x2 cloud lights, a B&W checkered rug, x3 reading lamps that could change multiple colors, a pastel purple mousepad and wrist rest, purple velour office chair covers, a colorful keyboard and mouse, and a mint green rolling cart that would hold my art supplies, which also came with small pencil holders which I planned on placing on each fold-out table. I also found a really cool DIY project someone had posted on Instagram that used a shadowbox and small figurines to make hanging decor.
Getting Supplies
Installing process
Day 1: I began the installation by assembling the rolling cart and trying out the seat covers I had found to confirm they were the right size.
Day 2: I assembled the pink tables, which worked perfectly. Although I determined that for the navigator, I should get a third pink table rather than slide the larger black picnic table under my main display table. I also installed the cloud lights and tested the mouse and keyboard capabilities on the large-screen display.
Day 4: I began hanging my string lights and tinsel, and also decided to lower the cloud lights slightly so they would be more noticeable. I also laid the rug out and tested my reading lamps and overall lighting. I laid out a tablecloth to decide between hot pink and silver, and ultimately landed on silver to break up the amount of hot pink used for the space’s backdrop. Lastly, I assembled my third pink folding table.
Day 5: On the fifth day, tragedy struck, and all my tinsel had fallen. With some help, I was able to get it back up, but it rattled me, as I still had so much to do on showcase preparations. I also began printing new participant templates to add to my pasteboard and decided which pictures to print to enhance my display. I settled on simple images of the original Misfit Dollz I had made and the 3D-printed counterparts from my past showcase. It was also suggested I make a sign-up sheet to pace out visitors.
Day 6 (the day of the showcase): The last step was to add the most important decorative elements, the very toys that make up the Wonderverse. I had determined a while ago that they should be incorporated by displaying them on the main table.
Final Preparations For the Activity & Promotion
For the activity component of this project, I cut out the sticker doll parts, printed my templates, and template covers on double-sided glossy paper, and bound the covers with colorful tape (as I was pressed for time). I made sure each activity table had a good balance of pencils and markers in their respective pencil holders, and, after receiving feedback, I purchased a small speaker so visitors could hear the music and sound effects as they navigated or viewed the virtual space. I also made two stickers using Canva, one to promote the project and one to promote my website.
Final Preparations For the Wonderverse
Making UI Features for The Main Menu & Instruction Page
Now that I had re-built the Wonderverse landscape, I could focus on creating UI elements for the main menu and instruction page. I had already begun developing text elements and a start page before my old file crashed, but I wanted to improve my original idea even more.
I would have two main buttons: (1) the start button, (2) the guide or movement manual button that would link to the instruction page. I used my signature Misfit Dollz background, which had been used in the Misfit Dollz commercial. I first used Unity’s UI elements to create buttons and panels to display my instruction page. However, this proved difficult, as panels or text would not stay anchored and would shift depending on screen size.
I determined that, rather than having separate UI vectors for my instruction page, I would insert an image that contained all of it. In terms of the instruction page’s design, I had two different ideas: (1) to create custom vectors that would represent my mouse and keyboard, where I would highlight relevant keys or mouse movements alongside descriptive text, or (2) to have an image of the very keyboard and mouse navigators would use, and have a diagram-style layout for instructions.
Ultimately, I landed on making my own UI vectors using shapes. I created a Photoshop document with the same dimensions as my UI canvas to avoid image distortion or vector shifts. After looking at different examples of how other games lay out their movement instructions, I then used simple shapes to create my vectors and arranged them on my Photoshop document. I used black and pink for my accent colors and white panels so my vectors would ‘pop’. For the header text, I used the same style as the title on the “Make Your Own Misfit Doll” template cover, which was created in Canva. I then created my vectors using black and pink accent colors and simple shapes. I looked at other examples of mouse-and-keyboard vectors that were effective in clearly conveying to the navigator how they can move in the world.
Adding Audio + Audio Editing
After collaborating with a friend on creating ‘Start page’ music, a soundscape for the main space, an ambient for my ‘Bear Scene’, and a small sound bite when interacting with 2D scans, I implemented them in Unity. I made an empty GameObject and attached an audio source component to play the starter music. I repeated this process for the background audio. For the ‘Bear Scene’ I repeated this process, but created a script to lower the volume of my main music and heighten the volume of the ambient track. I also made an audio trigger so that when the navigator enters the space, it triggers the script functions. I similarly did this in the Media Gallery; rather than simply triggering the script, it would also play the clips I had selected. I used the same audio trigger script to play the 2D scan soundbite when the navigator “collides” with a sprite, and I attached both the audio component and the script to all the pre-existing 2D scans.
Embedding Video
To embed my video clips in the Media Gallery, I decided to include three videos in total and create image textures to display scenes from the Misfit Dollz commercial drafts. The video on the ground floor would be an introduction video explaining to the navigator where they were located and what would be featured. Because I had a longstanding issue with my ‘elevator script,’ I decided to make the inconvenience part of the environment by having the audio guide explain how to move up floors if the elevator leaves without the navigator or acts erratically. I used Eleven Labs to create the audio track. I then added the audio to a Final Cut file, using a short clip of the 2D scans dancing in the Wonderverse, and overlayed a static TV visual effect to add a nostalgic element. The second embedded video was the Misfit Dollz commercial itself, since it was the main source of inspiration when initially concepting both the past Misfit Dollz project and the Wonderverse. The last embedded video was another narrative video, like the one before it, that jokingly explained how to get down if the elevator leaves you behind or doesn’t function properly. To embed a video, I had to create a video player element, a video rendering texture, and a special material to display the video. I then create a ‘video trigger’ script that used almost the same logic as the audio trigger script in the ‘Bear Scene’, it would (1) play the video and (2) lower the volume of the background music. While there were still some issues to work out, like the videos not pausing when leaving the Media Gallery floors or area, it still functioned, and with the allotted time, I felt it would be ok.
Last Minute Changes
Since I still had to manually remove the template background after the live demo templates were scanned and turned into sprites, I created a ‘demo reel’ video to give myself time to edit while keeping visitors entertained.
Week 13: Rebuilding the Wonderverse
This week, the worst-case scenario occurred: my file broke. What started as a lagging issue that had become gradually more noticeable over time had made my player controls unusable, the sprite movement glitchy, and, scariest of all, the console dialogue box in my Unity file was showing red error messages indicating that the lighting and URP pipeline were broken. I took a breath, considering various solutions to this issue, and came up with a plan. First, I checked previous scene drafts of the project, which, unfortunately, were also corrupted since they were in a shared project. Second, I tried using Unity version control, which also didn’t remedy the console messages or the lagging problem. By this point, I was eager to move on to creating UI features such as a main menu and an instruction page, embedding videos in the Media Gallery, and adding the music I had collaborated on with a friend, but there was no way to do that if my file was corrupted by constant errors. Somewhat defeated, I turned to Unity forums to see if there was any other means of salvaging the months of work I had done. With a glimmer of hope, there was one, the Unity Asset Manager, a built-in feature that would allow me to transfer assets from one project to another. Another factor I wanted to resolve was preventing lag in the new draft I would create, and I found that the reason my file was laggy was 3D assets with too high a poly-count. It felt like I was trying to save cargo from a sinking ship and needed to prioritize what I would ‘save’ and what was weighing me down. My new plan was as follows: (1) Identify which assets I currently have that are causing the most issues, (2) lower the poly-count of my 3D scanned assets on Meshy and Tripo, (3) use the asset manager to transfer the unproblematic assets.
Identifying Problematic Files & Building In Blender, and The Unity Asset Manager
The assets causing the most issues in the project in that current state were the tinsel grass assets and toy box shelves. What these two assets had in common was that I had repeatedly duplicated smaller assets to make up a larger one, and because I had added so many assets, the project lagged. I ultimately concluded that, rather than creating a larger 3D asset composed of multiple smaller assets, I should create a single asset containing all the same elements. I turned to Blender and began designing simple shelf assets using primitive shapes, and ‘joined’ those shapes to form one single asset in editing mode. I then went on Tripo, where I had first created the ‘Misfit Doll’ toy box assets, and reduced the poly count by using the ‘Retopology’ feature. In addition, I retextured these assets and deselected the ‘4K texture’ option to ensure they wouldn’t contain too much data. While the quality of the assets was lower, I felt it was important to have a functioning project over a few HD-quality assets.
I also went ahead and made additional boxes for the Misfit Dollz that hadn’t been added yet, using Artist to create the visuals and Tripo to create the 3D model. Following the creation and revision of the new boxes and the completion of my simple shelf asset, I used the Tripo bridge plug-in on Blender to seamlessly import the box models. After arranging the boxes on each shelf, I ‘joined’ the box assets to the shelf assets. I exported one of each shelf, in total 5 shelf assets, each containing one of the Misfit Doll boxes, as FBX files.
New Unity File & Asset Manager
I created a new project file using a more recent version of Unity and ensured it included both Unity Version Control and cloud save. First, I laid out the primitive planes for my main space, as I had in previous drafts. I then installed the Unity Asset Manager and imported my new shelf files from Blender. I began to transfer only what was necessary, such as spawning scripts, custom-made materials, and low-poly assets. Using the Unity Asset Manager, I attempted to transfer the already-made 2D sprites I had custom-rigged and animated, but to no avail. So I spent some time quickly re-rigging my scans and giving them new animations. For the 2D sprites that had shared rigs or were similar in shape, I copied/pasted animations to see which ones fit best. After transferring all the necessary assets, rigging, animating, and arranging my 2D sprites, I uninstalled the Asset Manager to lessen the project file size.
Building the Media Gallery and Narrative Structure, aka ‘Bear Scene’
Media Gallery
I then returned to Blender and began creating a new structure for the Media Gallery. I used the Tripo bridge to import the dollhouse asset and a Boolean function to cut out the shape of the house against a 3D plane. This would serve as the front of the media gallery. I then built the rest of the structure by scaling cubes for the walls and using another boolean function to create a rectangular cut-out for the floors that would allow the elevator to move through them. I made sure to merge the separate shapes into a single asset before exporting it from Blender as an .fbx file. In Unity, I adjusted each material to enhance its appearance by increasing the metallic and smoothing values. Lastly, I added the new elevator and attached the elevator script to it, though the elevator's functions proved finicky because the player would react to gravity, which confused the scripts. At this point, I wanted to move forward and focus on the next structure.
Bear Scene
For the ‘Bear Scene’, I took the bear tea party asset I had previously made on Meshy and imported it into Blender. I then made a floor and some walls and used a Boolean function to create cutouts with a tea-party plate asset I had 3D-rendered on Tripo, since I thought it was a cool shape. I then used Tripo’s image to 3D feature once more to generate 3D models of the toy tea set and toy soldiers I had sourced from earlier, and thought it would be interesting to build an archway entrance with those assets. Once completed, I laid out low-poly versions of existing assets I thought would fit my scene, including a toy lamp, a fireplace, and a grandfather clock. I ‘joined’ the structure into one asset and imported it into Unity. Using the Tripo and Meshy bridges, I added those same assets into the structure and placed it across from the Media Gallery.
Following these new developments, I laid out the remaining 3D models on my map as is and even included a low-poly tinsel block I had created in Meshy to diversify the terrain and tie my promotional poster to the world, as I thought it would make the project overall more cohesive. I even added some new items I had created using the image-to-3D feature on Tripo, such as a mini Furby and a toy soldier doing yoga that I had found at a joke shop.
Week 12: Ideating Installation Staging
Re-arranging my Spatial Layout
Initially, I was going to have my installation space nestled between a glass wall and a column within my assigned space. The layout would consist of a main display table, 2 smaller tables slightly lower, 2 office chairs, and a 55-inch TV to display the virtual space. At first at saw this layout as kind of a rabbit hole, a secretive, hidden magical space. However, logic set in, and I realized no one passing by would be able to see it, and they might not even know it’s there. So I decided to re-arrange. I first measured the total amount of space. Then thought of a few options as to how I could lay out my display table, chairs, and 55-inch TV.
I then met with a few peers to test my layout and show them my ideas for both the layout and the activity flow. One person suggested I have the ‘navigator’ of the virtual space at the head of the table, either slightly off to the left or right, so that visitors passing by could still see the large screen and decorations on the display table. I really liked this idea and went with it, re-arranging my initial furniture and set-up to match the description.
I noticed the yellow chairs I was using were a bit large and hard to adjust, so I switched them out for smaller office chairs that I would later decorate with seat covers. I felt that using one yellow chair would be perfect for the navigator; it is low enough that visitors could see over the navigator while still allowing the navigator to see the screen. I moved the office chairs closer to the head of the table so participants wouldn’t be too close to the screen, since it could hurt their eyes or be overwhelming.
Inspiration
I then began to think about my points of inspiration in terms of aesthetics, functionality, and what I would need in order to bring this vision to life. I looked at my old showcase, which I would reference as the basis for this new installation space. I also looked at external sources of inspiration, such as the set of PeeWee’s Playhouse, Nickelodeon bedrooms and gameshow sets, and the sets of Don’t Hug Me I’m Scared (DHMIS), and, most importantly, my initial inspiration: TeamLab’s ‘Sketch Ocean Installation’. It was now time to look at what materials and additional resources I would need to bring this to fruition.
Week 11: Drafting Template Cover & Making Custom Terrain Assets
Template Cover Development
When designing the template cover, I wanted to tie in both new elements developed for the Wonderverse while also nodding to the origin project that inspired it. I used the backdrop image from the Misfit Dollz commercial title page as the background, and created custom title text and instructions in Canva. At first, I was going to outsource my printing; however, price-wise, it wasn’t worth it. A bit perplexed and constrained by time, I decided to play into the DIY nature of the Wonderverse activity by handcrafting and hand-binding the template cover. I got double-sided glossy paper and made a front and back cover. For the front cover, I measured the most recent draft of my template to determine how large the cut-out window should be, which I would then cut out with an exacto knife. I formatted both covers using Photoshop and did a test print on regular printer paper. The artist duo from cirqueSaw, mentioned in a previous post, suggested that, for both the activity and the virtual space, I have multiple ways to display my instructions, and I kept this in mind when asking my friends and peers for feedback. One friend suggested I simplify and reformat my instructions to provide greater clarity, which I agreed with. Rather than a paragraph, I broke instructions down into steps. The color of the text boxes and text itself was chosen to match the aesthetic choices of the UI main menu page I was going to develop. Although I’m a maximalist, I wanted the text to be simple enough to be clearly legible, so I stuck with black text and pink text boxes, which, upon feedback from peers, was a good choice.
The Tinsel Grass Dilemma
Following the creation of my showcase promotional poster, I wanted to incorporate elements from it into my virtual environment. I went into Blender and created a particle system, specifically the one used to make “hair”. It looked great; however, upon exporting to Unity, my particle system did not transfer, just a blank plane asset. I then went online to remedy this issue and found that if I export the asset as an Alembic file, the particle system would transfer, but a new problem arose. It made this project lag even more than it already was, which I noted but didn’t address right away, something I would later regret. I then went into Unity and downloaded the ‘Terrain Assets’ package to see if I could apply my tinsel texture to a grass asset. While this worked, it wasn’t the look I was going for. I then turned to Tripo and Meshy to make tinsel “grass” to enhance the terrain and create a cohesive connection to the project's kitschy, whimsical elements. This proved more challenging than expected, and I didn’t get the results I was hoping for. When I got a Meshy generation I was ‘ok’ with, I duplicated the assets to cover the floors of my space, which at the time looked good, but again, my environment and player movement lagged more and more with each addition. For now, I kept it in place in hopes that the lag was due to being on Unity for an extended period of time. I needed to turn my focus towards my installation space.
Week 10: Expert Interview & Ideating Narrative-based Structures to the Virtual Environment.
Following my conversation with the artist duo, Nicole and Nathan, otherwise known as cirqueSaw, I focused more on developing my virtual environment. We discussed various topics, including audience engagement and audience capabilities within the virtual space. I asked about ways I could make implementing new scans in a live demo format more seamless while still keeping them engaging, posing two options I already had in mind: the first being a simple pre-made animation, such as bobbing or nodding, that didn’t require custom rigging, and the second being attaching a waypoint script to the scans, which also wouldn’t require custom rigging or animation. They said either option could work.
Testing New Spawning Script and Deciding between Method 1: Stationary Scans + Pre-made Animations or Method 2: A ‘waypoint ‘script
I referred back to Unity forums when creating a new spawning script. On a blank Unity scene, I added a basic floor and some staging assets. Then I first created multiple spawning points at the beginning of my map and also a rigged sprite of my bare template that each scan's texture or appearance would attach to. I then used this bare rigged sprite to create two animations. When working on the spawning script, I wanted the sprites to spawn randomly at these assigned points and randomly select one of the two animations I had created. At first, I planned to use both the animations and the waypoint script, but the waypoint script conflicted with the animations I created. When testing the ‘waypoint’ scripting option on its own, the scans looked strange as they moved across the map in a static t-pose, which ultimately led me to conclude that simple pre-made animations would work best. They still gave life to the scans, even when stationary, and blended more naturally with the custom-rigged, animated scans.
Lastly, following my interview with CirqueSaw, I decided that, rather than simply laying out my 3D toy assets as is, I should create structures that integrate the 3D models into the design and begin to think about what they would look like and where they would go. I returned to the image I had found of my childhood Beanie Babies having a ‘tea party’ or, as I joked, ’ritual sacrifice’, and wanted to incorporate that visual into my virtual space, leaving the ambiguity of its meaning to the viewer. Creating narrative-based 3D assets. I took my childhood image and used Tripo and Meshy to create a 3D model of the scene. When comparing, Tripo did not generate as good a result as Meshy, and the model also contained too many ‘faces’, so I ended up using the Meshy-generated 3D model and importing it into Blender to include in one of my new additions.
Week 9: Paper Doll Templates pt.II & Ideating Template Cover Design
When thinking about the designs I had made for my paper doll parts, I knew I wanted to diversify participants’ options and looked at the references I had already created. I thought about what I could add, such as a teddy-bear head or squid tentacles for legs, among others, and designed a few new options, for example, a ‘Sonny Angel Head’, ‘Robot Head’, and more options for ‘Hair’. While taking some time to further ideate on different paper doll parts, I also thought about how participants would assemble the parts and how they would stay on the template itself. Last week, I concluded that sticker paper would work best. For example, glue would get messy and may ruin the scanner, and tape would result in bumpy or uneven distribution. The conclusion I came to was that printing these parts on sticker paper would (1) allow for the pieces to lie flat on the template, (2) would not ruin the scanner, and (3) would be an easy application for those who choose this option.
Testing The Sticker Doll Parts
I had a friend test the paper doll parts printed on sticker paper. They made a few examples for me to scan, but noted the parts needed resizing. I kept this in mind as I revised and resized my doll part stickers. I really liked how they used various parts, for example, layering different limbs or using parts in places that they weren’t necessarily intended to be placed. I also noticed they cut pieces down to fit the template, which, while creative, may not be the most efficient in terms of design for those who just want to use the parts as is.
Ideating Template Cover Design & Creating the Template Reference Document
I also began to think about how I could more clearly convey the activity instructions to participants, especially considering the flow of the activity and my responsibilities or roles within the showcase. I wouldn’t necessarily be available to repeat myself on activity instructions if I were also editing the background out of scans or explaining virtual navigation. I had already done a rough prototype of a template cover, but I felt it was important to include written instructions directly on the template cover as a solution. Some feedback from my previous focus group sessions on this activity was to include examples or references to previous participants' designs. While I did make a poster board that I showed on ‘Demo Day’, I also thought about providing a document containing all the participant designs so far, which I would include alongside the template and template cover. I took the already completed scans and used Photoshop to create a double-sided layout. I felt that using a black background was best so that all the scans would be easily visible.
Week 8: Early Stages of Paper Doll Templates
Making Paper Doll Parts: The Artlist Method
This week, I focused on the activity template component of this project. I had received feedback following the first demo of the virtual world, staging, and activity suggesting I add paper doll cut-outs in addition to the blank template where participants could draw. I first thought about the outline of Mecha Barbz I had created on Artlist as a good test reference. I prompted Artlist to segment the parts of the character and then manually removed the background. I noticed when I asked it to segment Mecha Barbz that the hair would be a potential issue when adjusting the size of these segments into the template document, so I made two versions, Mecha Barbz’s head with hair and Mecha Barbz head without hair. This then led me to start creating different layers in Procreate with hair of varying lengths.
I organized the parts I have into different groups: (1) heads, (2) hair, (3) arms, (4) legs, and (5) bodies. For some of these parts, I used a combination of Artlist and Procreate; for others, I used only Procreate, tracing the outlines of different action figures or dolls for reference. The parts I created on Artist are from the original Misfit Doll designs, after repeating this process with my other Misfit Doll images, I noticed it would repeat parts that for from the reference image rather than the one I wanted to make parts for. In this case, I took parts that hadn’t already been replicated and made those paper doll parts as well. I prompted Artist to also replace the doll parts I created using the Misfit Doll images with entirely different parts. For example, using the newly created StringRay outline, I prompted it to replace the arms with bird wings, which worked well. I then cropped the rest of the image to show only the bird's wings and adjusted the size to fit the template's guidelines. I then pivoted my approach and decided to just trace existing reference images on Procreate.
Making Paper Doll Parts: Tracing Method with Procreate
I found an image of the transformer, ‘Optimus Prime’ and begun tracing it, making sure each limb I traced with on a different layer. Going forward, I plan to solely use this method.
At this stage, I am deciding (1) how large the variety of parts should be, (2) how many of each part should I print, (3) if hair and clothes should be their own categories separate from the heads or bodies, and (4) how detailed or simple these parts should be. Lastly, I plan to test the paper doll parts this week by printing them on sticker paper, arranging them on my template, scanning them, and rigging and animating them as I did for previous contributions. I am also holding an in-person group next week to test the paper doll version of this activity and design the activity template cover.
Week 7: Reflection & Thesis Development
This past week, I have (1) Debugged the 2D scan’s waypoint errors, (2) begun to experiment with terrain assets to enhance my environment, (3) made an updated early stage prototype for my ‘Make Your Own Misfit Doll Activity’, (4) conducted an expert interview with mixed media artist, and (5) Begun drafting my research paper, which led me down a moment of realization with regards to newly learned ways my project connects to the sources I’ve read.
Debugging The Waypoints & Improving the Lagging:
The inevitable crashing and lagging occurred during Demo day, a concern I wanted to solve as soon as I was able. I first checked to see what was causing this issue, questioning if it was the character controller? the camera? the render pipeline used in this version of Unity? or worst case scenario, a broken file. Luckily, after looking through various unity tutorials and forums, I found it was the character controller’s movement script and uneven navMesh that was causing the issue.
This quick fix that also allowed me to continue onward and remedy the ‘waypoints’ in which 2D scans would wander about the map. I created a new script and manually applied waypoints to each sprite. Now that I had 2D scans that moved about the map, I felt there was more interactivity and by solves these issues, I was able to turn my focus to my next goal, enhancing the materials and terraforming my environment.
Experimenting with Terrain Assets:
After receiving feedback on Demo Day and further reflection, I considered the layout of my environment. While there are elements of the space that are interactive, for example, the MisfitDoll box or HotWheelz assets are dynamic, I felt the layout was a bit flat, and from a bird's eye view (which one would have from the top floor of the Media Gallery), it appears as if my environment was floating in space. I wanted to create a space that was more exploratory by adding tunnels, stairs, ramps, and other elements that would make the space more playful. I thought of ways to achieve this goal aside from using pre-made assets or primitive 3D shapes. I began to make custom materials on Blender and also looked at the effects of different terrain brushes to adjust the height, create texture, and apply elements like water.
Activity Template Revisions:
Activity Template Revisions:
Following Demo day feedback, and survey results for remote and in-person participants, I found that there were cases in which individuals needed more guidance or may be less inclined to draw. A suggestion mentioned during my Demo day feedback was to “provide more options” and even brought up a previous idea I had been thinking about using in past iterations, paper dolls. I had previously held off on wanting to continue down the path of paper doll parts because I was concerned loosely glued pieces would get stuck in the scanner. I thought about sticky dots but they may make the surface of the paper uneven resulting in a lower quality scan. I then had the realization that I could use sticker paper. I would create my paper doll parts on Procreate, print multiple copies of the sketches, cut the individual pieces out and have them as options for participants to choose from. I am currently in the process of creating the paper doll parts, but I felt it was important first to look at various references from the real world. When speaking to my peers and friends about this project.
Writing & Drafting:
Writing & Drafting:
I primarily focused on delving into my literary sources and drafting my thesis. In the literature review, five major themes arose, (1) the cultural significance of toys, (2) the media effects of toy marketing and commercials, (3) toys role in art & activism, (4) creative play personalization & The DIY movement, and (5) nostalgic practices & ‘kidulting’ as therapeutic practices. When thinking of how these themes have influenced my prototype ideas following more recent iterations. I looked back at some of my old toys, and reached out to family to ask if they had any photos of the toys I used to play with. One caught my eye in particular, which was a 2002 photograph of what appears to be a beanie baby tea party. As an adult revisiting this image made me chuckle because it reminded me a bit of a summoning, and I realized I was actively re-contextualizing that memory and those objects in a new lens.
Expert Interview
Following this realization, I also conducted an expert interview with Adios From Everywhere, a mixed-media artist who primarily works with visual, audio, digital, and installation art, and received helpful feedback as well as visual, audio, and literary references that could help develop it further into a more engaging and exploratory experience. We primarily discussed staging and subversive imagery, as I have also begun brainstorming different media clips featuring the Misfit Dolls or reimagining the ones created for the origin project using Mixamo to animate and Unity to stage. We also discussed the activity template upon showing them the early development of my updated prototype. I explained how I was making a folder-like cover that I would design in Procreate and print on vinyl to display instructions, enhance the component's aesthetic function as part of the project, and provide participants with past examples or references from other collaborators. They suggested that I use cellophane on the cutout portion of the cover to resemble a toy box, an idea I really liked. I also showed them the photograph I had found earlier that week, and they found it amusing, suggesting that I should incorporate it into my virtual world, perhaps by creating a hidden room that the ‘player’ has to discover.
Next Steps:
All these changes and realizations have helped me move forward in terms of how to truly transform and elevate this project, by adding discoveries or hidden passageways, incorporating media from my own childhood experiences, as the photograph mentioned previously, and the paper doll and template cover additions. Since I did pivot to focusing on writing a bit more this week, I plan on doing the following for the final stretch of this project:
- Tweaking animations to make 2D scans smoother.
- Making sure waypoint paths don’t result in 2D scans colliding.
- Creating my paper doll template pieces on Procreate.
- Creating my template cover on Procreate.
- Adding hidden compartments and walkable structures of different heights and shapes.
- Enhancing the environment with terrain elements.
- Collaborating with a friend who makes music to create soundscapes and effects for assets and the environment.
- Collecting images of my childhood toys and incorporating Misfit Dollz content to create experimental media clips.
Week 6: Demo Day Draft, Prep, & Feedback
Drafting The Landscape + Structures
The week leading up to Demo Day, I focused on (1) further developing my environment, (2) gathering participants and collecting participants' templates for my 'Make Your Own Misfit Dollz" Activity, (3) animating, rigging and integrating participant scans, (4) prepping and staging for Demo Day. This progress and preparations and the setbacks they came with in some cases, lead to breakthroughs with regards to environment, 'game feel', the 2D scanning process and the activity component of my thesis itself.
When building my environment, began to use Unity's built-in AI Navigation system in which I could have sprites travel between waypoints while still acknowledging and not overrriding the structures and landscape's shape and materials. I also began to design and create the structure for the Media Gallery, which has since been further ideated upon following Demo Day. Using my basic layout of primitive 3D planes, I began to think about where I would place participants scans within the MisfitDollz: Wonderverse (rebranded), as well as which 3D models I would have as static obstacles and which I would have as dynamic or moving obstacles. I decided the Misfit Dollz Boxes at the starting points should be dynamic and react to physics as well as the Hot Wheelz to make the environment more interactive.
After placing these assets on my layout, I created a Navmesh surface that would inform playable 3D assets or 2D Participant Scans of which surfaces are walkable, like the ground or the media gallerie's elevator and which are not walkable, such as walls and some static 3D assets that act as world boundaries.
Activity, Collaboration, and Participant Scan Integration
Activity Templates
Following these developments, I reached out to participants to either contribute in-person or remotely. Ideally this activity would be done in-person where I would be able to provide context on both the thesis and the project, however I felt it important to collect enough contributions to present or simulate the participants' scans function and interactivity within the newly developed environment. I asked particpants to time themselves to test how long on average the activity would take, and sent a brief survey to provide feedback on their experience. Overall, participants really enjoyed this activity, some even making more than one Misfit Doll to contribute. Some feedback I received from participants, especially those who worked on the activity remotely was that there should be more guidance or structure in terms of what to create or how, which helped inform the development of a new template prototype. The guidance ranged from wanting more instruction to visual examples of what other particpants have created and so I also though of ways to showcase participant contributions for both the showcase and for collaborators.
Scanning, Rigging, & Animation
After collecting up to 17 participant scans, I begun the process of integrating, rigging, and animating, the scans into my Unity project. I first opened a PNG in Preview and removed the background. I then converted the scans into Photoshop format or .PSDs. Using Unity's .PSD exporter, I created a custom present to format the scans into 2D sprites. In the Sprite editor I used the rigging and animation features to (1) created a custom shape that would react to physics or movement, (2) created a rig or bones so move the limbs of the 2D scans, (3) generated weights for each limbs to determine the speed and dominance of the movement, (4) experimented with custom animations to bring the drawings to life. The scanning/rigging process would range from 3 minutes to 3 hours with some drawings being particularly finnicky do to the colors that were used, drawing outside the lines, or not using enough color at all. This realization made me think about how I would present my live demo and I decided that perhaps I would apply the static 3D sprites to invisible 3D planes that move between waypoints rather than a fully articulated sprite. I then placed the scans around my map and added a particule effect to enhance their appearance, I also give them Rigidbodys so if the collided with either by the 'player' or another sprite they would respond to the physics. While this idea worked in theory the inevitable crashing of my laptop occurred and resulted in a broken navigation mesh, an issue I have since solved on my newest draft.
Staging Prep
To stage my Demo Day space I thought it was important to provide background context by including the original Misfit Doll sculptures, the toys that were 3D scanned and placed in the virtual world, a display of the original commercial, a posterboard of participants printed out scans to showcase the collaborative contributions and provide new participants with references. I set up my Unity Project on a large monitor and laid out markers, pencils and templates for participants to use.
Demo Day + Feedback
While there were some technical setbacks, I felt the feedback I received on my project gave me new inspiration on ways I can make the activity more streamlined or clear to participants. The feedback provided included suggestions to provide more artistic options for the activity such as adding the original paper doll concept to my template as well as the drawing version and similar sentiments on providing participants with more guidance on the activity. When I asked about what to include if anything to the space I developed on Unity, feedback suggested I focus more on developing the media gallery and refining the assets I already have. From this feedback I have prototyped a new template, begun working with Unity built-in character controls for smoother (not laggy movement) and terrain shaders and assets for more refined environment building. I also thought about my layout and staging, with concerns my space may be too busy, so I have begun reworking the staging as well. At this stage, I am primarily focused on created content to embed into my media gallery and the structure of my thesis paper & research.
Next Steps
While the layout of the environment was coming together, I encountered some errors upon creating this new draft:
- Integrate new terrain elements into the Misfit Dollz: Wonderverse and make sure the navMesh and waypoints are functional.
-
Create pre-made animations and skeletons for 2D scans to streamline the Template -> Animated Sprite pipeline.
-
Develop and complete my new template prototype.
-
Create Misfit Dollz media content to embed in the media gallery (including images and video made with the 3D-rigged, animated Misfit Doll models).
-
Re-organize/Re-design staging.
- Collect more participant scans to include in the Wonderverse and more survey data.
Demo Day Gallery:
Week 5: More Photogrammetry, Environment Building, & Brainstorming Interactivity(originally published 03/02/2026)
This week, I focused primarily on the concepts and research that complement this thesis project. Keeping in mind the notion that toys are "cultural signifiers," I went to Jackpot Deals and searched for toys that I could then 3D scan into 3D assets that would then be placed in my Unity environment. The toys I collected included plastic army men, Hot Wheels, little figurines, and other small toy accessories. I used both Polycam & Tripo to scan my objects and then converted these scans into FBX & OBJ files.
Starting Space + Asset Development
After scanning most of the objects collected at Jackpot Deals, I began creating Boxes for the MisfitDolls themselves. Since I already had an 'AllexBear' box as a reference, I went into Artlist, prompted Nano Banana to replace the toy in the box with the toy in the second image reference, and prompted it to customize colors that complement the figure. I did this for both SonnyGearSolid & MechaBarbz, and am currently in the process of making boxes for MD Queer & StringRay (although this is a lesser concern)
Then, using the shelf asset I made in an earlier draft, I began to design a space in which the participant or 'player' would begin their journey. I then added the new MisfitDollz Box assets to the shelves, coordinating the box assets between each character design.
Designing the Main Space
Following the development of the starting space, I decided that the center plane or main space, would be where the 'player' could interact with the 3D toy assets, interact with the Misfit Dollz 3D models, and trigger an interaction between MechaBarbz (or another misfit doll) narration that would explain the space and the concepts behind its design. The player then has two areas that can be explored, to the left, a Media Gallery and to the right 'StringRay's Surf Mini Game'.
Media Gallery
The media gallery would function as a space where current and past Misfit Dollz media will be displayed. For example, I plan on embedding the already made MisfitDollz Commercial onto a 3D plane, as well as the past concept/poster art, and newly created animation/video art created using, Mixamo to animate the MisfitDoll models, Final cut for editing the video content, and Procreate to make captions or other graphics. I may design it to look similar to my showcase back in December, though I am still thinking about that portion of the layout.
StringRay Surf 'Mini Game'
When you approach StringRay's area it will trigger an interaction where the 'player' will be told that they need to collect missing doll parts (this would involve a simple script for collectibles) so that Stringray has a clean beach to surf on. I included this as an idea since it is a potential mini interaction and StringRay's persona is based around surfing and the beach (the reason being that wearing swimsuits and beach attire is a huge anxiety-inducing for most people, let alone queer or trans individuals who are experiencing body dysmorphia or dysphoria). As a sidenote: Each original Misfit Doll has an anecdote or design that informs that character's backstory/development. If this isn't an interactive game, it may simply be an extension of the environment, but adding mini-games or small tasks could make this project more interactive.
Setbacks...Compromises?
While the layout of the environment was coming together, I encountered some errors upon creating this new draft:
- Input-Manager Errors: The input settings to 'look' or rotate the camera kept on having issues, It would give me an error operation or not function at all. I am currently in the process of correcting this error by troubleshooting and changing the 'Input manager' component.
- The Participant Scans: While I was able to add and render the scans in the older file, I wanted to improve how the scans are rendered, interacted with, and displayed. I came across some issues while recreating the older method on the new file, as a solution, I am considering either having the Participant scans render in a new scene (separately) or finding a plug-in or library for real-time rendering and PNG background removal.
Planning This Week:
Technical
- Finish the 'paper doll' templates on Procreate and time the process on creating and scanning vs. the drawing templates.
- Experiment with different workflows for the Participant Scans including plug-ins or libraries.
- Finish scanning all the toys/objects acquired for the environment and collectibles.
- Solve the input manager settings so that I can rotate my camera or 'look' while in motion.
- Put my animated Misfit Dolls into the new project file and add 'navmeshes'.
- Begin designing the Media Gallery and choosing media to embed.
Conceptual
- Continue to work on my thesis research.
- Begin to design the interactive experiences ranging from more passive interactions to event triggers.
- Create media content for the virtual gallery.
Week 4: Photogrammetry & New Environment- (originally published 02/23/2026)
This week, I focused on building my Misfit Dollz Universe in Unity and began to plan my thesis paper. For world-building, I decided to redo the layout and change the perspective from 3rd-person to 1st-person, since I felt it was more immersive. The animated Misfit Dollz models would be a part of the environment rather than the "player". I also began scanning the items from the previous showcase to add as 3D models to the Unity project. The items consist of toys I had collected from dollar stores, ranging from 'make-believe play' props, such as a cash register, a toy doctor set, and knock-off Power Ranger figures. Something that came to mind as I was scanning my objects is their design and the objects themselves. One concept that I plan on further exploring in my research is how these items are marketed and how they perpetuate and reinforce gender conditioning, and how this can result in negative self-image and distorted views on gender and gender expression.
Toys serve as cultural and social signifiers, and having symbolism, such as the toy mentioned previously, introduces harmful body image that has a lasting negative psychological impact on individuals into adulthood, especially within the queer community, where such products are usually exclusive and limited in the gendered expressions they embody. Another example that comes to mind is how baby dolls were marketed to young girls in order to condition them to motherhood, or how the design of an Easy-Bake and its purpose were also directed towards this concept of gendered products and domesticity. While the symbolism and marketing found in these toys and accessories can reinforce negative cultural or gender stereotypes, there is a light at the end of the tunnel; many adults, especially queer adults, have revisited their childhood toys and collect them as a form of therapy or healing. Some view it as a means of reclaiming the narrative surrounding toys and their symbolism. Community and nostalgia are core aspects of this project, as the purpose of this thesis is not only to address the issue of gender-typing in toys in a satirical and experimental way, but also to create a community that is healing together through their shared experience and an activity that engages their self-expression and joy.
Back to Progress!
I first created a new space and designed my own assets for the shelves lining the entry to the space. Using Polycam, a photogrammetry app, I then took some 3D scans of my objects and uploaded them to Tripo to refine and correct them. I used Multi-image --> 3D and exported them into Unity. I then began placing them in my space as a part of the environment. I also thought about what objects I just want as a part of the space and what objects I want to interact with. I know that I plan on adding a 3D model of a retro TV with the Misfit Dollz commercial embedded to add context to the project, but I still need to see what interactions I actually want in the space. To get more guidance and feedback, I will be asking my peers as well as friends or acquaintances who fall under the primary demographic this project serves.
Moving Forward
Moving forward this week, I plan to scan more objects, collect more objects from dollar stores that I feel embody the symbolism addressed in my research, and add the Sprite Rendering & Scanning functions to this new Unity file. In terms of the participant scans and sprite rendering, I also want them to render in the main environment to connect the two components of the installation. Perhaps, rather than people drawing, I could cut out parts for paper dolls that people can mix and match. I think this may be a better approach because it won't take as long as a drawing if I have people participating live, and it may also better align with the project's aesthetics.
Week 3: Paper Templates, Scanning, & Rendering - (originally published 02/17/2026)
Set-Up & Troubleshooting
I took a break from developing the virtual environment for the MisfitDollz: XR Experience and began testing scanning and rendering of the 'Make Your Own Misfit Doll' template prototype. I created three examples to test with a small portable scanner and, using the scanner's compatible software, I confirmed that the new scans automatically relocated to the 'Scans' folder in my Unity project. Since my plan is to remove the background to show only the sprite, I explored both options in the scanning application and in Unity to execute this successfully.
Scanning -> Rendering
After some research, I learned I could create a script that would render all the scans in my project folder and also crop and remove the background from my template. Last week, I experimented with attaching the sprite to a particle system, but found that if I had multiple sprites, they would need to be on the same texture atlas to render, which would take too long to format or edit in real time. I instead opted to create an empty GameObject parent and a 'child' to render the 2D visual. To have my scans spawn properly, I created a button that would prompt the scans to appear on Display 2, a camera to show the renders, an empty game object to locate the spawns in my scene, and another empty gameObject to carry out the functions of the background removal and spawning script. I also created a simple animation script to add some movement to the scans. It's also important to note that I purposefully made a multi-colored example, a monochromatic example, and a grayscale example to see how the background removal and cropping script would affect the scans when rendered. While this was somewhat successful, there are still some issues to address. There were bits of the original border still attached to the new sprite; the alpha or 'white background' removal threshold worked well for the colorful and monochrome examples, but not for the grayscale. The renders tend to overlap each other, and I am still trying to figure out ways to adjust these settings. While the newer scans would relocate to my Unity project in real time, they would not render to the project when it was running in real time.
I still, however, consider this progress. I was able to have my scans render, have the background removed, have my renders shown on a second display, now it was a matter of polishing and cleaning up these renders, and adding more interactive elements to the rendered scans.
Next Steps
My next steps following this weeks prototype testing would be (1) to polish the rendering settings for the scans, (2) find a way for the newly added scans to render in real time while the program is running, (3) finish the main Unity environment, (4) think of what interactivity I want in the Misfit Dollz Unity environment, (5) clean up animation transitions for 3D models, make a "Choose Your character" start page for the Misifit Dollz Universe component of this project. I also plan on gathering a focus group to test my 'Make Your Own Misfit Dollz' activity.
Week 2: Worldbuilding & AR - (originally published 02/09/2026)
Brainstorming
When thinking about my installation and the interactions participants and viewers have with the project, I considered ways to engage a community by facilitating a live participatory performance. TeamLab has an installation called 'Sketch Ocean' in which participants are given a template with an outline of a sea creature. They can color in or design their sea creature within the outline's boundaries. The template is then scanned, and the scan is applied as a texture that wraps around a 3D shape or skeleton, rendering in real time. It is then displayed on another screen. I was inspired by this idea and looked into how I could make my own version for my 'Make Your Own Misfit Doll' activity. Since I am already developing a Misfit Dollz world, I think it would be interesting to see whether I can either render participants' creations in the world I created or display them as a UI overlay in front of the scene. You can have someone exploring the world and this would be the background of the display, 2 forms of interaction. I also thought of ways I can make the world itself more interactive. I thought of smaller activities that can be accessed by passing through a portal and entering a new scene.
ArUco Markers
I looked up how to create an installation like teamLab's Sketch Ocean and found that, if I use OpenCV for Unity, I can generate my own ARUCO markers for the outlines so the program knows which model should appear. The problem is being able to not only render the model, but more importantly, the custom-drawn texture from the outline. This is where I am struggling. I tested the OpenCV plug-in, and while the markers I generated worked, I am somewhat lost on how to apply the texture. Another issue I came across was that if people were to choose paper-cutout parts and have them assembled, it would be hard to indicate which model should be rendered. A solution I considered was simply making the outline a vague humanoid silhouette, so it would use only one simple model. Another solution would be to make a "create your own misfit doll (avatar)" scene in Unity where people can choose parts from different 3D assets.
A Potential Solution?
When investigating ways I can add my 2D scanned images to my Unity project, I tested the idea in a few ways: (1) I applied the 2D B&W outline to a quad game object by creating a custom material, (2) I made a custom particle system that would be able to render custom shapes. The particle system solution may be an effective route to take. I would just like to see how I can quickly add these textures to my Unity file while it is running in real time!
With this potential solution, I could get closer to my goal of adding participants' art to the Misfit Dollz world. I will draft a few different templates and generate ARuco codes assigned to empty quads (which solves the issue of too many meshes to choose from and saves time). I will then see if there are scripts or workflows that will allow me to attach these scans to the 3D quad without having to do it manually. IF this isn't possible, I will find a quick manual method that would be executed at the showcase.
What I Can Control & Worldbuilding
Since I am very new to OpenCV and struggling a bit with it, I decided to spend more time on the Unity environment I had started building. I used the MechaBarbz model that I had rigged last week as a test avatar. I created a basic player movement script so that the avatar can move around the space using the arrow keys. I built the space (so far) using basic starter assets (shelves, ground) as well as my own custom assets I developed in Tripo (Misfit Doll Boxes). I used a Galaxy skybox asset to further customize the environment to make it feel surreal and match the elements of the commercial.
Next Steps
What I plan on doing this week:
- Finish building the primary environment
- Add interactivity to the environment
- Make a "Choose Your Avatar" start page (for roaming )
- Learn more about AR & arUco markers/Particle systems and 2D texture wrapping
- Make a working template/develop an ARUco library/ find quick import/render methods for the 2D scans
- Test the 'Make Your Own Misfit Doll' component with a small sample group (2-3 people)
Week 1: The Pivot & The Plan - (originally published 02/02/2026)
The Pivot
Initially, I was going to develop a thesis project based on the research I had done in Pre-thesis about hybrid learning and immersive educational experiences, specifically a mixed-media, XR project on the Scale of Cells and the Electromagnetic spectrum. However, I decided to pivot for three main reasons, (1) Feasability, while my original thesis project had a lot of research backing and supporting my original hypothesis, I found that developing and designing such a project over 3 months may not be the most practical as well as having already developed aspects of my new thesis project, (3) Passion & Motivation, while I think immersive, multi-sensory and accessible learning is important, Misfit Dollz: The Wonderverse is a project I am emotionally connected to and I feel in the current time we are living in, especially with the portrayal of beauty and aesthetics distorting our self-perception on digital platforms, an uprise in “conservative values”, christian nationalism, and queer oppression, I think this project serves an important purpose.
Background & Context
Misfit Dollz: The Wonderverse is an extension of a previous project that focuses on topics surrounding queerness, standards of beauty, specifically gender dysphoria & body dysmorphia, gender conditioning in marketing, art activism & subversive art, and self-perception. It serves as a satirical critique of how toy design, commercial, and media portrayals perpetuate gender stereotypes and unrealistic standards of beauty marketed towards children. This project was inspired by the work of the Barbie Liberation Organization (BLO), as well as Tom Forscythe, who made a series called ‘Food Chain Barbie’ that famously won a case against Mattel. This project is also inspired by my experience as a queer person navigating my identity under the heteronormative gaze and influence of these commercials and toys, and their lasting negative psychological impact on mental health and self-image.
Misfit Dollz: The Exhibit - Commercial Draft (2025)
Project Idea
For this project, I would like to develop an XR experience made up of two primary installations: (1) A game in which the view can navigate a virtual space that takes on the elements of the Misfit Dollz commercial and expands the Misfit Dollz universe, this would include 3D scanned props from my showcase, my commercial embedded in a 3D asset as well as environmental elements, (2) an AR face/body/object tracking photo booth (disguised as a mirror) in which a person can capture themselves with a custom Misfit Dolls filter (this involves AR elements such as animated 3D model overlays). I think both can serve not only to develop the thesis but also to world-build.
The space, overall, would evoke a child's bedroom, with a kitschy aesthetic to match the Misfit Dollz maximalist aesthetic. I chose this spatial concept because it simulates a nostalgic, familiar space that also feels safe and private, where a lot of self-reflection occurs.
Target Audience
The target audience would be primarily adult queer or non-cisgender identifying individuals who may have also had negative experiences brought about by the standards of gender and beauty set by the media or enforced by their childhood toys. I want this project not only to address the harm of gender conditioning in toys & commercials, but also to create a space where queer people can come together and find others who may resonate with the work.
Research Topics of Interest
The areas of study that I plan to investigate to further inform this concept and develop it into a thesis would be (1) Gender Studies, (2) Behavioral & Social Psychology, (3) Activism & Subversive Art, (4) Media Studies, commercial influence in particular, (5) Queer Studies, (6) Kitsch, Camp and other related styles or aesthetics (think Sontag’s ‘Notes on Camp’). Research methods I plan to utilize for this project would be (1) Archival Research surrounding the topics of interest previously stated, viewing old commercials, toy designs, gender studies etc., (2) Cultural Probes, where I design a kit in which the targeted demographic comment on their experience with gender dysphoria & body dysmorphia as well as their relationship or citing of a childhood toy or commercial that may have evoked those feelings, (3) Observational, creating a focus group of queer individuals and observing participants interaction with a selection of mainstream name brand toys, followed by a survey that asks what toys they were most drawn to, which ones made them uncomfortable, and other notable thoughts they may have on the project and activity overall, (4) Activity-based, where I would develop an activity and template in which a group of participants would design their own Misfit Doll and discuss their ideation process in a follow-up group conversation & feedback survey. Lastly, I would have a final group to playtest both the AR filter & Unity game experience, and take a survey to provide qualitative & quantitative feedback.
What Has Been Done Previously (Not for Thesis)
- Sourced toys and physical materials
- Dissembled & re-assembled doll parts
- 3D scanned using Qlone & polished reference images using Chat GPT (background clean-up)
- Further polished 3D models in Tripo
- Issue: Tripo’s rig feature would not work on my models—>Pivoted to making an AI-generated commercial.
- Generated images, video for the commercial + posters using Midjourney, Artlist, Chat GPT, LTX studio
- Generated song using Suno AI, narration + sound effects using Eleven Labs
- Printed version of 3D models (though this was troublesome as the models’ textures were being finicky), set up a file using Blender)
- Edited commercial on Final Cut & Veed (for background removal and green screen)
- Designed & staged the Exhibit space
- Gave a small live performance in which I gave viewers a sticker of their choice (Misfit Dollz sticker) from a toy cash register as a playful nod to the consumerism associated with name-brand dolls overall.
What I Want To Do (For Thesis)
- So far, I’ve found a workflow to finally rig my models for animation. These animated models would apply to both the AR and 3D Unity installations.
- Tripo—> Meshy—>Mixamo—> Tripo —> Blender & Unity
- Create an AR file with newly rigged 3D models on Touch Designer or Unity
- Think about the interactions and design plan for AR installation
- So far, I have begun sketching the AR experience
- Using Unity, make a 3D environment in which the participant navigates the space as one of the Misfit Dollz (or encounters them)
- Design & Develop probe kits, activity templates, and surveys to conduct the research methods stated above.
- Begin Archival Research by identifying archives that hold artifacts and documents related to the area of study I plan to investigate.
- More research to inform my thesis paper & project (ideally 15-20 sources for literature review).
- Design & Create Unity environment, character actions, and plan interactivity (3D world installation)
- Design & Create TouchDesigner AR installation
- Decide what it will look like & how people can interact with it
- Incorporate object tracking, perhaps with 3D printed objects
- Make sure to test the number of people the AR filter can identify (to see how many people can use it at once)
- Design and construct physical displays
- Incorporate posters into AR elements (when someone uses their phone camera, they can view an animation over the poster or a pop-up that provides a background story for each character and how they were concepted).