pix2pixHD

Synthesizing and manipulating 2048x1024 images with conditional GANs

View on GitHub↗Upstream NVIDIA/pix2pixHD↗

Builder

NVIDIA

NVIDIA • big-tech

Stars

6,970

Using upstream star count

Forks

1,417

Using upstream fork count

Open Issues

Activity Score

0/100

0 commits in 30d

Created

Dec 1, 2017

Project creation date

README Summary

NVIDIA's pix2pixHD is a PyTorch implementation of high-resolution image-to-image translation using conditional GANs. The model can synthesize photorealistic images at 2048x1024 resolution from semantic label maps and supports interactive object editing. It builds upon the original pix2pix framework with architectural improvements for generating high-quality, detailed images.

AI Dev Skills

Unmapped

Tags

Generative Adversarial NetworksConditional Image GenerationHigh-Resolution Image SynthesisImage-to-Image TranslationComputer VisionDeep Learning Architecture DesignMulti-scale Discriminator NetworksFeature Matching LossAdversarial TrainingOn-premiseUrban PlanningImageData Augmentation for Computer VisionGamingSelf-hostedArchitectureVirtual Environment CreationAutonomous VehiclesHigh-Resolution Image EnhancementConditional GenerationSemantic Segmentation to Photorealistic Scene GenerationLabel Map to Street Scene SynthesisGenerative AIHigh-Fidelity Content GenerationSketch to Photograph ConversionDigital ArtEntertainmentPython

Recent Activity

Updated 1 years ago

7 Days

30 Days

90 Days

Quality

research

Quality: medium
Maturity: research

PM Skills

Product DiscoveryUser Experience

Languages

Python100.0%

Timeline

Project created: Dec 1, 2017
Forked: Mar 14, 2026
Your last push: 1 years ago
Upstream last push: 1 years ago
Tracked since: Nov 4, 2024

Similar Repos

pgvector cosine similarity · $0

Loading…