Skip to content

Dataset · Annriki123

Poster and SVG pairs

Annriki123/Poster2SVG

Corpus of 20,600 pairs of poster images and corresponding SVG files for editable reconstruction.

What's inside

Created for the PosterSVG project, containing 20,600 pairs split into 20,000 training and 600 test samples. The data was gathered from public web pages, filtered, and cleaned. Each sample consists of an input poster raster image (input.png) and a corresponding SVG file (gt.svg).

What you can do with it

  • Training models for vector reconstruction and editing of poster images
  • Researching raster-to-SVG conversion
Languages
Chinese, English
Size
10K–100K rows
Downloads this month
515
Likes
3

How to load it

from datasets import load_dataset

ds = load_dataset("Annriki123/Poster2SVG")