# thunil/TecoGAN

This repo contains source code and materials for the TEmporally COherent GAN SIGGRAPH project.

Repository: https://github.com/thunil/TecoGAN
Canonical: https://ross.abutalabs.com/products/tecogan
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2023-08-17T20:55:56+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2734, "days_push": 1112, "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 6141, forks 1131 (observed 2026-08-28T04:09:36.469876+00:00)

## What it is
TecoGAN is the official source code for a temporally coherent GAN for video super-resolution, published at SIGGRAPH/ACM TOG. It includes inference, training, evaluation metrics, and pre-trained models built on TensorFlow.

## Use cases
- upscale low-resolution videos with AI
- generate temporally coherent video super-resolution
- train a GAN for video enhancement
- evaluate video super-resolution with temporal metrics
- download pretrained video upscaling models

## When to choose
- you need research-grade video super-resolution with temporal coherence
- you want to reproduce or extend the TecoGAN paper
- you have an Nvidia GPU and TensorFlow setup

## When to avoid
- you need a simple production-ready video upscaler with easy installation
- you cannot use TensorFlow 1.x or an Nvidia GPU
- you need active development or recent feature updates

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, deep-learning, video-processing, image-processing
- domain: computer-vision, deep-learning, image-processing
- platform: python
- tags: video-super-resolution, gan, tensorflow, temporally-coherent, pretrained-models, research-code, video, linux, gpu

## Member repositories
- thunil/TecoGAN (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:36.469876+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 2026-08-29T17:47:34.311264+00:00, confidence not recorded.
  - readme: https://github.com/thunil/TecoGAN (fetched 2026-08-28T04:09:36.469876+00:00, sha 7ece8acc342f)
- Data as of 2026-08-30T08:39:29.467469+00:00.