search

google-research-datasets / hiertext

The HierText dataset contains ~12k images from the Open Images dataset v6 with large amount of text entities. We provide word, line and paragraph level annotations.
Creative Commons Attribution Share Alike 4.0 International
261 stars 23 forks source link

metric problem #15

Closed MelosY closed 1 year ago

MelosY commented 1 year ago

Could you please report the precision,recall of word detection and paragraph detection in paper "Towards End-to-End Unified Scene Text Detection and Layout Analysis"?

Jyouhou commented 1 year ago

Sure: paragraph: (P/R/F/Tightness/PQ) 76.04, 62.45, 68.58, 78.17, 53.60

word: 69.57, 82.68, 75.56, 78.04, 58.97

MelosY commented 1 year ago

Sorry to bother you again，Could you please report the precision,recall，pq of word detection ,line and paragraph detection on validation set？

Jyouhou commented 1 year ago

Sure! Happy to help.

It's actually right in this repo: https://github.com/google-research-datasets/hiertext#sample-output-on-the-validation-set