david-gpu/srez

Health v2 · maintenance only

Flags: no_releases archived

Adoption not part of the score

5270 stars · 654 forks observed · 2026-08-28

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

A deep learning project that performs 4x image super-resolution on 16x16 images using a DCGAN-based architecture with ResNet generator modules, trained on the CelebA face dataset. It is a research/demo codebase written in Python with TensorFlow.

Use cases

• upscale low-resolution face images with deep learning
• train a GAN for image super-resolution
• generate plausible 64x64 faces from 16x16 inputs
• experiment with DCGAN architectures in TensorFlow
• reproduce super-resolution results on the CelebA dataset

When to choose

• you want a compact, readable reference implementation of GAN-based super-resolution
• you are working with well-lit, frontal face images similar to CelebA
• you want to learn how adversarial plus L1 loss improves upscaling

When to avoid

• you need a maintained, production-ready super-resolution tool
• you need general-purpose image upscaling beyond frontal faces
• you need modern TensorFlow 2.x or PyTorch compatibility

Facets

library · maturity abandoned

machine-learning deep-learning image-processing llm-training machine-learning computer-vision image-processing deep-learning python windows super-resolution gan dcgan tensorflow face-images celeba image-upscaling gpu linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="david-gpu/srez")