Titan AI LogoTitan AI

CodeFormer

17,496
3,640
Python

Project Description

[NeurIPS 2022] Towards Robust Blind Face Restoration with Codebook Lookup Transformer

CodeFormer: [NeurIPS 2022] Towards Robust Blind Face Restoration with Codebook Lookup Transformer

Project Title

CodeFormer — Robust Blind Face Restoration with Codebook Lookup Transformer

Overview

CodeFormer is an open-source Python project that focuses on robust blind face restoration using a Codebook Lookup Transformer. It stands out for its ability to enhance old photos, fix AI-arts, and restore faces in images and videos, leveraging a novel approach in the field of face enhancement and restoration.

Key Features

  • Codebook Lookup Transformer for robust blind face restoration
  • Face enhancement and restoration capabilities
  • Face inpainting and colorization features for cropped and aligned face images
  • Support for video input enhancement
  • Online demo integrations with platforms like Hugging Face, Replicate, and OpenXLab

Use Cases

  • Use case 1: Enhancing old photos and restoring faces in historical images
  • Use case 2: Fixing AI-generated art that has face-related imperfections
  • Use case 3: Improving the quality of faces in video content for better viewer experience

Advantages

  • Advantage 1: State-of-the-art technology for blind face restoration
  • Advantage 2: Supports a wide range of input types, including images and videos
  • Advantage 3: Open-source and actively maintained with regular updates and new features

Limitations / Considerations

  • Limitation 1: Requires specific hardware (Pytorch, CUDA) for optimal performance
  • Limitation 2: May have limitations in handling faces with extreme occlusions or distortions

Similar / Related Projects

  • Project 1: FaceSwap - A project focused on real-time face swapping, but does not specialize in restoration.
  • Project 2: DeepFaceLab - A tool for creating deepfakes, which can be used for face manipulation but not specifically for restoration.
  • Project 3: OpenCV - A comprehensive computer vision library that includes face detection but not the specific restoration capabilities of CodeFormer.

Basic Information

Requirements:

  • Pytorch >= 1.7.1
  • CUDA >= 10.1
  • Other required packages in requirements.txt

📊 Project Information

  • Project Name: CodeFormer
  • GitHub URL: https://github.com/sczhou/CodeFormer
  • Programming Language: Python
  • ⭐ Stars: 17,395
  • 🍴 Forks: 3,618
  • 📅 Created: 2022-06-21
  • 🔄 Last Updated: 2025-08-20

🏷️ Project Topics

Topics: [, ", c, o, d, e, b, o, o, k, ", ,, , ", c, o, d, e, f, o, r, m, e, r, ", ,, , ", f, a, c, e, -, e, n, h, a, n, c, e, m, e, n, t, ", ,, , ", f, a, c, e, -, r, e, s, t, o, r, a, t, i, o, n, ", ,, , ", p, y, t, o, r, c, h, ", ,, , ", r, e, s, t, o, r, a, t, i, o, n, ", ,, , ", s, u, p, e, r, -, r, e, s, o, l, u, t, i, o, n, ", ,, , ", v, q, g, a, n, ", ]


🎮 Online Demos

  • [Hugging Face
  • [Replicate
  • [OpenXLab
  • [OpenXLab
  • [Hugging Face
  • [Replicate

📚 Documentation


This article is automatically generated by AI based on GitHub project information and README content analysis

Titan AI Explorehttps://www.titanaiexplore.com/projects/codeformer-505667511en-USTechnology

Project Information

Created on 6/21/2022
Updated on 9/17/2025