What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis. – NLPIR自然语言处理与信息检索共享平台

自然语言处理与信息检索共享平台 自然语言处理与信息检索共享平台

What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis.

NLPIR SEMINAR Y2019#41

INTRO

In the new semester, our Lab, Web Search Mining and Security Lab, plans to hold an academic seminar every Monday, and each time a keynote speaker will share understanding of papers on his/her related research with you.

Arrangement

Tomorrow’s seminar is organized as follows:

  1. The seminar time is 1:20.pm, Mon (December 16, 2019), at Zhongguancun Technology Park ,Building 5, 1306.
  2. Wang Gang is going to give a presentation on the paper, What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis.
  3. The seminar will be hosted by Changhe Li.

Everyone interested in this topic is welcomed to join us.

What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis.

Jeonghun Baek, Geewook Kim, Junyeop Lee1 Sungrae Park, Dongyoon Han, Sangdoo Yun, Seong Joon Oh, Hwalsuk Lee

Abstract

Many new proposals for scene text recognition (STR) models have been introduced in recent years. While each claim to have pushed the boundary of the technology, a holistic and fair comparison has been largely missing in the field due to the inconsistent choices of training and evaluation datasets. This paper addresses this difficulty with three major contributions. First, we examine the inconsistencies of training and evaluation datasets, and the performance gap results from inconsistencies. Second, we introduce a unified four-stage STR framework that most existing STR models fit into. Using this framework allows for the extensive evaluation of previously proposed STR modules and the discovery of previously unexplored module combinations. Third, we analyze the module-wise contributions to performance in terms of accuracy, speed, and memory demand, under one consistent set of training and evaluation datasets. Such analyses clean up the hindrance on the current comparisons to understand the performance gain of the existing modules. Our code is publicly available.

You May Also Like

About the Author: nlpvv

发表评论