Text is Text, No Matter What: Unifying Text Recognition using Knowledge Distillation
release_z2ezlupwyrh5xjykeh6bdkjwg4
by
Ayan Kumar Bhunia, Aneeshan Sain, Pinaki Nath Chowdhury, Yi-Zhe Song
2021
Abstract
Text recognition remains a fundamental and extensively researched topic in
computer vision, largely owing to its wide array of commercial applications.
The challenging nature of the very problem however dictated a fragmentation of
research efforts: Scene Text Recognition (STR) that deals with text in everyday
scenes, and Handwriting Text Recognition (HTR) that tackles hand-written text.
In this paper, for the first time, we argue for their unification -- we aim for
a single model that can compete favourably with two separate state-of-the-art
STR and HTR models. We first show that cross-utilisation of STR and HTR models
trigger significant performance drops due to differences in their inherent
challenges. We then tackle their union by introducing a knowledge distillation
(KD) based framework. This is however non-trivial, largely due to the
variable-length and sequential nature of text sequences, which renders
off-the-shelf KD techniques that mostly works with global fixed-length data
inadequate. For that, we propose three distillation losses all of which are
specifically designed to cope with the aforementioned unique characteristics of
text recognition. Empirical evidence suggests that our proposed unified model
performs on par with individual models, even surpassing them in certain cases.
Ablative studies demonstrate that naive baselines such as a two-stage
framework, and domain adaption/generalisation alternatives do not work as well,
further verifying the appropriateness of our design.
In text/plain
format
Archived Files and Locations
application/pdf 1.0 MB
file_tb6jld2svbggbdx5fzzp27e6e4
|
arxiv.org (repository) web.archive.org (webarchive) |
2107.12087v2
access all versions, variants, and formats of this works (eg, pre-prints)