# XPixelGroup/HAT

CVPR2023 - Activating More Pixels in Image Super-Resolution Transformer TPAMI - HAT: Hybrid Attention Transformer for Image Restoration

Repository: https://github.com/XPixelGroup/HAT
Canonical: https://ross.abutalabs.com/products/hat
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2024-06-02T11:55:42+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1589, "days_push": 822, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1583, forks 185 (observed 2026-08-28T04:05:07.249963+00:00)

## What it is
HAT (Hybrid Attention Transformer) is a PyTorch implementation of a state-of-the-art transformer model for image super-resolution and restoration, published at CVPR 2023 and extended in TPAMI. It includes pretrained models, real-world GAN-based SR variants, and inference tooling with tile mode for limited GPU memory.

## Use cases
- upscale low-resolution images with a transformer model
- restore real-world degraded photos to high resolution
- benchmark super-resolution models on Set5, Set14, Urban100, Manga109
- train a custom image super-resolution model
- sharpen and denoise images with GAN-based SR

## When to choose
- you need top-tier image super-resolution quality
- you want pretrained models for 4x upscaling or real-world SR
- you're doing research on transformer-based image restoration

## When to avoid
- you need fast, lightweight upscaling on CPU or mobile
- you only need simple bicubic or Lanczos resizing
- you lack a GPU and can't tolerate slow inference

## Facets
- artifact type: library
- maturity: stable
- function: image-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, deep-learning
- platform: python
- tags: super-resolution, transformer, image-restoration, cvpr2023, pytorch, gpu, linux

## Member repositories
- XPixelGroup/HAT (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:07.249963+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:55:59.936489+00:00, confidence not recorded.
  - readme: https://github.com/XPixelGroup/HAT (fetched 2026-08-28T04:05:07.249963+00:00, sha 05edb6358bd6)
- Data as of 2026-08-30T08:39:29.467469+00:00.