Skip to content
Antoine Boucher

Options

Profiles

← All articles
Article

Make figurines with AI for 3D printing

4 min read

Listen to the audio story

From a photo to a resin figurine · 1:56 · AI-written script, synthetic voices

Transcript

Ava: Here's a trick. Take a photo, and a few steps later you're holding a little vinyl-style figure of it, printed in resin.

Andrew: Okay, that sounds like it should be harder than it is. How does it actually work?

Ava: Antoine wrote it up as a four-step pipeline. First, a reference photo, a clear side or front view. Second, turn it into a chibi vinyl toy image. Third, an image-to-3D model builds the mesh. Fourth, you print.

Andrew: So the AI does two jobs. It restyles the picture, then it builds the 3D shape.

Ava: Exactly. For the toy image he uses ChatGPT, Midjourney, or ComfyUI with a model called OmniGen2. The prompt is simple: a vinyl chibi figurine, smooth plastic, white background, one character, in a pose that can stand on a base.

Andrew: And the 3D part?

Ava: Hunyuan3D version 2. In his ComfyUI workflow, one graph does both stages. The first stage makes the chibi picture, the second outputs two mesh files at slightly different thresholds. And if your graphics card only has eight to twelve gigabytes of memory, he says to lower a couple of resolution settings.

Andrew: Does it matter much which 3D tool you pick?

Ava: Less than you'd think. The page says the clarity of the toy image matters more than the brand of the generator. Mesh tools amplify a bad silhouette, so a blurry start gives you a mushy figure.

Andrew: Fair. And printing?

Ava: Resin, for small parts. Tilt overhangs toward the plate, keep supports light on things like hair and capes, and check for floating islands in the slicer or in UVtools. Skip that check and a failed print can waste hours.

Andrew: So the takeaway is: spend your effort on the 2D image.

Ava: Yes. Get the picture right, and the rest of the pipeline mostly follows. The full write-up has the diagram, the prompts, the workflow file and the slide deck.

Make figurines with AI for 3D printing
Table of contents

A clear 2D toy image gives the image-to-3D model a useful starting point. Follow the steps with the slide deck and ComfyUI workflow.

1. Pipeline (in order)

  1. Reference photo — clear side or front view.
  2. Chibi vinyl render — ChatGPT, Midjourney, or ComfyUI (OmniGen2).
  3. Mesh — Hunyuan3D v2 (local ComfyUI or hosted tools).
  4. Print — orient, supports, resin slice; fix islands with UVtools[1] if needed.

Photo to Object — photo, AI toy image, 3D model, resin print

Pipeline sequence (Mermaid source):

sequenceDiagram
    autonumber
    actor User
    participant Photo as Reference photo
    participant ComfyUI
    participant OmniGen2 as OmniGen2 stage
    participant Hunyuan3D as Hunyuan3D v2 stage
    participant Slicer as Resin slicer
    participant Printer

    User->>Photo: Side or front reference
    User->>ComfyUI: Load chibi workflow + reference-photo.jpg
    ComfyUI->>OmniGen2: ReferenceLatent + chibi vinyl prompt
    OmniGen2-->>ComfyUI: PNG (white background)
    ComfyUI->>Hunyuan3D: CLIPVision + image-to-mesh
    Hunyuan3D-->>ComfyUI: GLB (VoxelToMesh 0.7 / 0.6)
    ComfyUI-->>User: mesh/ComfyUI*.glb
    User->>Slicer: Orient, supports, hollow
    Slicer-->>User: Sliced job
    User->>Printer: Resin print
    Printer-->>User: Chibi figurine

Deck (~17 MB) — screenshots and tool URLs from the talk:

Download slide deck (.pptx) Open PDF ~17 MB
Make figurines using AI for 3D printing · slide 1
Make figurines using AI for 3D printing · slide 2
Make figurines using AI for 3D printing · slide 3
Make figurines using AI for 3D printing · slide 4
Make figurines using AI for 3D printing · slide 5
Make figurines using AI for 3D printing · slide 6
Make figurines using AI for 3D printing · slide 7
Make figurines using AI for 3D printing · slide 8
Make figurines using AI for 3D printing · slide 9
Make figurines using AI for 3D printing · slide 10
Make figurines using AI for 3D printing · slide 11
Make figurines using AI for 3D printing · slide 12
Make figurines using AI for 3D printing · slide 13
Make figurines using AI for 3D printing · slide 14
Make figurines using AI for 3D printing · slide 15

2. Step 1 — 2D toy / chibi

Deck prompt:

Convert this image into a toy-style plastic figure. Keep features but simplify and smoothen as if molded in plastic.

Before / after Toy result

Chibi prompt (same idea, bigger head):

Vinyl chibi figurine, smooth plastic, white background, one character, pose that can stand on a base.

Gallery (ChatGPT):

Bernese puppy Golden retriever Cowboy Pool Spider

3. Step 2 — ComfyUI (one graph)

Download: image_omnigen2_image_edit.json — Load in ComfyUI (OmniGen2 chibi edit + Hunyuan3D v2 mesh). Put your photo in input/ as reference-photo.jpg (or edit the LoadImage node).

StageWhat it does
AOmniGen2 + ReferenceLatent → chibi PNG (heun, 45 steps)
BHunyuan3D v2 → two GLBs (mesh/ComfyUI*.glb, thresholds 0.7 and 0.6)

Full positive/negative prompts and seeds are in the JSON. Needs OmniGen2 + Hunyuan3D-2 custom nodes (Hunyuan3D-2[2]). On 8–12 GB VRAM, lower octree_resolution or num_chunks.

4. Step 3 — Image to 3D (hosted)

Convert to 3D with Hunyuan3D and Rodin

Single-image mesh tools — clarity of the toy PNG matters more than the brand.

ToolLinkNotes
Hunyuan3D-2GitHub[2]Open weights / local or hosted runs; strong general image-to-3D.
RodinHDWebGL demo[3]High-detail avatars and figurines from images.
Hi3DGenHugging Face demo[4]2D diffusion prior + geometry consistency (CHI 2024 paper[5], white paper PDF[6]).
Hyper3Dhyper3d.ai[7]Fast commercial image-to-3D.
Meshymeshy.ai[8]Text/image to 3D with editing and remeshing.
Tripo3Dstudio.tripo3d.ai[9]Browser studio workflow.
Sparc3Dlizhihao6.github.io/Sparc3D[10]Structured sparse-view reconstruction.

Tripo3D studio

Hyper3D

Sparc3D

Meshy AI

Hi3DGen (CHI 2024[5], demo[4]) — strong when the 2D silhouette is already clean.

5. Mesh preview

Sample GLB files from Hunyuan3D, RodinHD, Hyper3D, and Hi3DGen are not stored in this repo (each was 20–75 MB). Use the generator previews in the tool UIs, or export from your own ComfyUI run (mesh/ComfyUI*.glb).

Mesh preview in generator UI

6. Step 4 — Mesh settings

Generator settings (animated)

Mesh preview (animated)

Settings detail

Settings panel

Tune face count, symmetry, and remesh in the generator UI before export.

7. Step 5 — Resin print

Resin print setup

Printed figurine on build plate

Resin for small parts. Tilt overhangs toward the plate; light supports on hair and capes.

Islands in slicer

Check floating islands with the slicer or UVtools[1] before printing.

8. Printed results

Finished figurine 1

Finished figurine 2

9. When to use this pipeline

GoalStart here
Gift figurine from a pet photoChatGPT/Midjourney chibi → hosted mesh
Repeatable local batchComfyUI JSON in this bundle
Teaching a classSlide deck and pipeline diagram
Production CAD interchangeExpect cleanup in Blender after GLB

10. Pitfalls

  • Skipping 2D quality — mesh tools amplify bad silhouettes.
  • Committing 50 MB GLBs — export locally; link tools instead.
  • Zero island check — resin failures mid-print waste hours.

12. References

References

  1. sn4k3. (). UVtools. [Software repository] github.com.
  2. Tencent-Hunyuan. (). Hunyuan3D-2. [Software repository] github.com.
  3. WebGL demo. (). [Web page] rodinhd.github.io.
  4. Hugging Face demo. (). [Web page] huggingface.co.
  5. CHI 2024 paper. (). [Web page] doi.org.
  6. white paper PDF. (). [Web page] stable-x.github.io.
  7. hyper3d.ai. (). [Web page] hyper3d.ai.
  8. meshy.ai. (). [Web page] www.meshy.ai.
  9. studio.tripo3d.ai. (). [Web page] studio.tripo3d.ai.
  10. lizhihao6.github.io/Sparc3D. (). [Web page] lizhihao6.github.io.

References use this article’s publication date.