正在加载视频...
视频加载失败
All these consistent 3D characters were made with AI in under 20 min via a fast, simple workflow 🧵👇 🛠️ One tool: - Custom LoRA - Flux Kontext (back view control) - 3D gen: Rodin / Tripo / Hunyuan ⏱️ Get clean meshes + PBR textures in record time.
10 条评论
Step 1 - "Front View" Generation Start by generating the front view of each character (with a custom LoRA to lock in the style) I used “Action Plastic", a model designed to create characters with the toy-like aesthetic of classic action figures Link 👉
Step 2 - "Back View" Once the front view is ready, generate the back. Head over to and select Flux Kontext ✏️ Prompt: “Generate the back view of this character” (feel free to add details) This step gives you control over front/back before going 3D.
Step 3 - Click “Convert to 3D” on any character you like. Choose your preferred AI model from 7 different generators (Rodin, Tripo, Hunyuan, etc.). For better accuracy, load the back view from in Step 2. Then just hit Generate, and get an accurate textured 3D mesh in minutes.
Step 4 - Compare Models Instantly If not happy with the first result: 🔁 Switch the 3D model in one click - no need to re-upload anything or switch app. 🆚 Generate outputs side by side to compare mesh and texture fidelity. Quick, flexible iteration to find the best fit.
Step 5 – Export the full collection Once your your character set is ready: 📥 Batch select and download all your 3D assets at once into Blender or any 3D software of your choice for further editing. From idea to a full pack of editable 3D assets... in under 20 minutes.
+ Enjoy automatic indexing of all assets, smart search and filters (type, author, date, model, collection, tags…) - so you can find past work in seconds, even months later. SOC2 compliant. SSO/SAML & enterprise features available. More details 👇
Still looks like uniform topology though. Probably good for 3D printing and not much else?
AI meshes are dense, it's no secret 🫣. You can tweak mesh density in the model settings - down to 500 faces or less with Hunyuan 2.1 on simple objects, for ex. For more complex assets, bring them into Blender for cleanup/retopo (until AI retopology makes it faster & easier)
wishing there will be a proper workflow for lowpoly models soon 🤞
How about these?
![✨ Made a new mini feature on Photo AI: [ Grab from 3d model ] So the problem is we're at that stage in time (typical for AI) where image-to-3d models are not good enough but are fun to play with, but we know they'll be good enough in 1-2 years With [ Make 3d model ] you already can turn any Photo AI pic into a 3d model but it still looks hyper clunky and deformed, but it works! One cool idea I had to make that more useful and made now: Let people make a 3d model then change the view of the it with the 3d viewer, then press [ o ] and it grabs a frame of the 3d That image you can then [ Remix ] (img2img), and it becomes a real photo again and that in turn you can then turn into a video again with [ Make video ] So that essentially gives you a fully freeform camera position control to take photos with One thing I need to fix is the background/skybox, I kinda need to take the original photo and remove the person and just get the background for the 3d model viewer, in this case it should be white, but it's a start!](https://image.24vids.com/tw-1940749265615179946/amplify_video_thumb/1940747973064822784/img/P___fqKPPJyKeVwq.jpg)
