dunbar12138/pix2pix3D

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1716 stars · 149 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

pix2pix3D is the official PyTorch implementation of a CVPR 2023 paper on 3D-aware conditional image synthesis. It generates 3D objects (neural fields) from 2D label maps such as segmentation or edge maps, with an interactive 3D editing demo.

Use cases

• generate 3D objects from 2D segmentation maps
• synthesize photorealistic images from edge maps with controllable viewpoints
• edit 3D content interactively from any viewpoint
• train a 3D-aware conditional GAN on monocular image and label map pairs
• render semantic meshes from generated neural fields
• convert face segmentation maps into 3D-consistent face images

When to choose

• you need viewpoint-consistent 3D generation conditioned on 2D label maps
• you want an interactive tool for 3D-aware image editing
• you are reproducing or building on the CVPR 2023 3D-aware conditional synthesis paper
• you need pre-trained models for seg2face, seg2cat, or edge2car tasks

When to avoid

• you need a production-ready application with a polished UI
• you lack a CUDA-capable GPU, since training and inference are compute-intensive
• you need general-purpose 3D modeling rather than label-map-conditioned generation
• you need actively maintained code with frequent updates and support

Facets

library · maturity maintenance

machine-learning deep-learning image-processing computer-vision graphics computer-vision graphics machine-learning artificial-intelligence python 3d-gan neural-radiance-fields conditional-generative-model pix2pix eg3d cvpr-2023 pytorch research-code 3d-editing neural-fields linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="dunbar12138/pix2pix3D")