Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

DOES SUPER-RESOLUTION IMPROVE OCR PERFORMANCE IN THE REAL WORLD? A CASE STUDY ON IMAGES OF RECEIPTS

Abstract: 

Recently, many deep learning methods have been used to handle single image super-resolution (SISR) tasks and often achieve state-of-the-art performance. From a visual point of view, the results look convincing. Yet, does it mean that those techniques are reliable and robust enough to be implemented in real business cases to enhance the performance of other computer vision tasks? In this article, we investigate the use of SISR to construct higher resolution images of real receipt photos sent by a company’s customers and evaluate its impact on the performance of an OCR task (receipt information retrieval). Using built in task-based performance evaluation methods, we show that the use of SISR can significantly improve OCR performance in the case where recognition was poor in low resolution, but can also deteriorate the performance for receipts that were already successfully recognized. As a conclusion, we provide recommendations on how to best use SISR in a production environment.

up
0 users have voted:

Paper Details

Authors:
Vivien Robert, Hugues Talbot
Submitted On:
13 November 2020 - 12:52pm
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Vivien Robert
Paper Code:
2624
Document Year:
2020
Cite

Document Files

ICIP Thesis Presentation.pdf

(12)

Subscribe

[1] Vivien Robert, Hugues Talbot, "DOES SUPER-RESOLUTION IMPROVE OCR PERFORMANCE IN THE REAL WORLD? A CASE STUDY ON IMAGES OF RECEIPTS", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5553. Accessed: Nov. 29, 2020.
@article{5553-20,
url = {http://sigport.org/5553},
author = {Vivien Robert; Hugues Talbot },
publisher = {IEEE SigPort},
title = {DOES SUPER-RESOLUTION IMPROVE OCR PERFORMANCE IN THE REAL WORLD? A CASE STUDY ON IMAGES OF RECEIPTS},
year = {2020} }
TY - EJOUR
T1 - DOES SUPER-RESOLUTION IMPROVE OCR PERFORMANCE IN THE REAL WORLD? A CASE STUDY ON IMAGES OF RECEIPTS
AU - Vivien Robert; Hugues Talbot
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5553
ER -
Vivien Robert, Hugues Talbot. (2020). DOES SUPER-RESOLUTION IMPROVE OCR PERFORMANCE IN THE REAL WORLD? A CASE STUDY ON IMAGES OF RECEIPTS. IEEE SigPort. http://sigport.org/5553
Vivien Robert, Hugues Talbot, 2020. DOES SUPER-RESOLUTION IMPROVE OCR PERFORMANCE IN THE REAL WORLD? A CASE STUDY ON IMAGES OF RECEIPTS. Available at: http://sigport.org/5553.
Vivien Robert, Hugues Talbot. (2020). "DOES SUPER-RESOLUTION IMPROVE OCR PERFORMANCE IN THE REAL WORLD? A CASE STUDY ON IMAGES OF RECEIPTS." Web.
1. Vivien Robert, Hugues Talbot. DOES SUPER-RESOLUTION IMPROVE OCR PERFORMANCE IN THE REAL WORLD? A CASE STUDY ON IMAGES OF RECEIPTS [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5553