mingyuliutw/UNIT

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

2025 stars · 358 forks observed · 2026-08-28

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

UNIT is NVIDIA's research implementation of unsupervised image-to-image translation networks (NIPS 2017), translating images across domains like day-to-night or snowy-to-summery without paired data. It is a research codebase superseded by the Imaginaire reimplementation and licensed CC BY-NC-SA 4.0 for non-commercial use.

Use cases

• translate images between domains without paired training data
• convert day photos to night scenes
• change weather in street scene images
• translate between dog breeds or cat species
• reproduce NIPS 2017 image translation paper results
• convert synthetic images to realistic ones

When to choose

• you need the original UNIT method for research reproduction
• you want pretrained models like synthia-to-cityscapes translation
• you're studying unsupervised GAN-based image translation

When to avoid

• you need a maintained or performant implementation - use NVlabs/imaginaire instead
• you need commercial licensing - the CC BY-NC-SA 4.0 license is non-commercial
• you need multimodal or many-to-many translation - use MUNIT instead
• you need Python 3 support - the code targets Python 2.7

Facets

library · maturity maintenance

machine-learning deep-learning image-processing deep-learning computer-vision image-processing python gan image-translation pix2pix nvidia research-code pytorch linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="mingyuliutw/UNIT")