# yulunzhang/RCAN

PyTorch code for our ECCV 2018 paper "Image Super-Resolution Using Very Deep Residual Channel Attention Networks"

Repository: https://github.com/yulunzhang/RCAN
Canonical: https://ross.abutalabs.com/products/rcan
Language: Python
License Family: other
Last push: 2025-04-01T05:43:22+00:00

## Health v2 (maintenance only)
Score: 39/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 14, release rhythm 35, longevity 100
- inputs: {"age_days": 2983, "days_push": 519, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1503, forks 318 (observed 2026-08-28T04:04:54.685904+00:00)

## What it is
PyTorch implementation of RCAN, a very deep residual channel attention network for single image super-resolution from an ECCV 2018 paper. It provides training and testing code built on the ESR-PyTorch codebase, plus pretrained visual results for scales 2, 3, 4, and 8.

## Use cases
- upscale low-resolution images with a deep learning model
- train a super-resolution network on DIV2K
- reproduce RCAN PSNR/SSIM benchmark results
- compare channel attention architectures for image restoration
- get pretrained super-resolution models for scales 2-8

## When to choose
- you need a well-cited reference implementation of channel attention for super-resolution
- you want to reproduce ECCV 2018 paper results or build on RCAN architecture
- you have a CUDA GPU and want to train or test SR models in PyTorch

## When to avoid
- you need a maintained production library with a license and active support
- you want a simple CLI upscaler without setting up training data and PyTorch 0.4-era dependencies
- you need super-resolution on CPU or in a browser

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, image-processing, deep-learning
- domain: computer-vision, image-processing, deep-learning
- platform: python
- tags: super-resolution, pytorch, channel-attention, eccv-2018, research-code, linux, gpu

## Member repositories
- yulunzhang/RCAN (main) score 39

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:54.685904+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 2026-08-30T04:32:50.944817+00:00, confidence not recorded.
  - readme: https://github.com/yulunzhang/RCAN (fetched 2026-08-28T04:04:54.685904+00:00, sha f32e819279fe)
- Data as of 2026-08-30T08:39:29.467469+00:00.