# whai362/PSENet

Official Pytorch implementations of PSENet.

Repository: https://github.com/whai362/PSENet
Canonical: https://ross.abutalabs.com/products/psenet
Language: Python
License: Apache-2.0
License Family: permissive
Topics: icdar2015, total-text, ctw1500, psenet
Last push: 2023-04-07T06:52:32+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3021, "days_push": 1244, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1186, forks 338 (observed 2026-08-28T04:03:55.140830+00:00)

## What it is
Official PyTorch implementation of PSENet, a progressive scale expansion network for robust scene text detection. It includes training, testing, and evaluation scripts for benchmarks like ICDAR 2015, Total-Text, and CTW1500.

## Use cases
- detect text in natural scene images
- train a scene text detection model on ICDAR 2015
- evaluate text detection on Total-Text and CTW1500 benchmarks
- extract text regions from photos before OCR
- benchmark PSENet against other text detection methods
- fine-tune a pretrained text detection checkpoint

## When to choose
- you need a proven scene text detection model with published benchmark results
- you want the official reference implementation of the PSENet paper
- you work with PyTorch and standard text detection datasets
- you need pretrained checkpoints for ICDAR 2015, Total-Text, or CTW1500

## When to avoid
- you need text recognition (OCR) rather than detection only
- you want a maintained production framework - the project recommends MMOCR instead
- you need Python 2 compatibility or very recent PyTorch versions
- you require commercial support or frequent updates

## Facets
- artifact type: library
- maturity: maintenance
- function: computer-vision, ocr, machine-learning, deep-learning
- domain: computer-vision, image-processing, machine-learning, artificial-intelligence
- platform: python
- tags: scene-text-detection, psenet, pytorch, icdar2015, total-text, ctw1500, research-code, linux, gpu

## Member repositories
- whai362/PSENet (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:55.140830+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 2026-08-30T06:24:05.719958+00:00, confidence not recorded.
  - readme: https://github.com/whai362/PSENet (fetched 2026-08-28T04:03:55.140830+00:00, sha d76da2663438)
- Data as of 2026-08-30T08:39:29.467469+00:00.