NVIDIA/pix2pixHD

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

6923 stars · 1412 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

PyTorch implementation of pix2pixHD, a conditional GAN method for synthesizing and manipulating high-resolution (2048x1024) photorealistic images from semantic label maps. It supports image-to-image translation tasks like label-to-streetview and label-to-face, plus interactive editing via instance-level manipulation.

Use cases

• generate photorealistic street scenes from semantic label maps
• synthesize portrait images from face label maps
• translate low-resolution segmentation maps into 2k images
• interactively edit objects in generated images
• reproduce CVPR 2018 pix2pixHD research results
• train a conditional GAN for image-to-image translation

When to choose

• you need high-resolution (1k-2k) semantic image synthesis with GANs
• you want the reference implementation of the pix2pixHD paper
• you need label-to-photo translation for Cityscapes-style or face datasets

When to avoid

• you need modern diffusion-based image synthesis quality
• you lack an NVIDIA GPU with 11GB+ memory
• you want a production-ready, actively maintained image generation pipeline

Facets

library · maturity maintenance

machine-learning deep-learning image-processing computer-vision deep-learning computer-vision image-processing machine-learning python gan pytorch image-to-image-translation image-synthesis conditional-gan research-code semantic-image-synthesis linux macos gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="NVIDIA/pix2pixHD")