Cargando…

Brickognize: Applying Photo-Realistic Image Synthesis for Lego Bricks Recognition with Limited Data

During the last few years, supervised deep convolutional neural networks have become the state-of-the-art for image recognition tasks. Nevertheless, their performance is severely linked to the amount and quality of the training data. Acquiring and labeling data is a major challenge that limits their...

Descripción completa

Detalles Bibliográficos
Autores principales: Vidal, Joel, Vallicrosa, Guillem, Martí, Robert, Barnada, Marc
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9967933/
https://www.ncbi.nlm.nih.gov/pubmed/36850495
http://dx.doi.org/10.3390/s23041898
_version_ 1784897388923584512
author Vidal, Joel
Vallicrosa, Guillem
Martí, Robert
Barnada, Marc
author_facet Vidal, Joel
Vallicrosa, Guillem
Martí, Robert
Barnada, Marc
author_sort Vidal, Joel
collection PubMed
description During the last few years, supervised deep convolutional neural networks have become the state-of-the-art for image recognition tasks. Nevertheless, their performance is severely linked to the amount and quality of the training data. Acquiring and labeling data is a major challenge that limits their expansion to new applications, especially with limited data. Recognition of Lego bricks is a clear example of a real-world deep learning application that has been limited by the difficulties associated with data gathering and training. In this work, photo-realistic image synthesis and few-shot fine-tuning are proposed to overcome limited data in the context of Lego bricks recognition. Using synthetic images and a limited set of 20 real-world images from a controlled environment, the proposed system is evaluated on controlled and uncontrolled real-world testing datasets. Results show the good performance of the synthetically generated data and how limited data from a controlled domain can be successfully used for the few-shot fine-tuning of the synthetic training without a perceptible narrowing of its domain. Obtained results reach an AP50 value of 91.33% for uncontrolled scenarios and 98.7% for controlled ones.
format Online
Article
Text
id pubmed-9967933
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-99679332023-02-27 Brickognize: Applying Photo-Realistic Image Synthesis for Lego Bricks Recognition with Limited Data Vidal, Joel Vallicrosa, Guillem Martí, Robert Barnada, Marc Sensors (Basel) Article During the last few years, supervised deep convolutional neural networks have become the state-of-the-art for image recognition tasks. Nevertheless, their performance is severely linked to the amount and quality of the training data. Acquiring and labeling data is a major challenge that limits their expansion to new applications, especially with limited data. Recognition of Lego bricks is a clear example of a real-world deep learning application that has been limited by the difficulties associated with data gathering and training. In this work, photo-realistic image synthesis and few-shot fine-tuning are proposed to overcome limited data in the context of Lego bricks recognition. Using synthetic images and a limited set of 20 real-world images from a controlled environment, the proposed system is evaluated on controlled and uncontrolled real-world testing datasets. Results show the good performance of the synthetically generated data and how limited data from a controlled domain can be successfully used for the few-shot fine-tuning of the synthetic training without a perceptible narrowing of its domain. Obtained results reach an AP50 value of 91.33% for uncontrolled scenarios and 98.7% for controlled ones. MDPI 2023-02-08 /pmc/articles/PMC9967933/ /pubmed/36850495 http://dx.doi.org/10.3390/s23041898 Text en © 2023 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Vidal, Joel
Vallicrosa, Guillem
Martí, Robert
Barnada, Marc
Brickognize: Applying Photo-Realistic Image Synthesis for Lego Bricks Recognition with Limited Data
title Brickognize: Applying Photo-Realistic Image Synthesis for Lego Bricks Recognition with Limited Data
title_full Brickognize: Applying Photo-Realistic Image Synthesis for Lego Bricks Recognition with Limited Data
title_fullStr Brickognize: Applying Photo-Realistic Image Synthesis for Lego Bricks Recognition with Limited Data
title_full_unstemmed Brickognize: Applying Photo-Realistic Image Synthesis for Lego Bricks Recognition with Limited Data
title_short Brickognize: Applying Photo-Realistic Image Synthesis for Lego Bricks Recognition with Limited Data
title_sort brickognize: applying photo-realistic image synthesis for lego bricks recognition with limited data
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9967933/
https://www.ncbi.nlm.nih.gov/pubmed/36850495
http://dx.doi.org/10.3390/s23041898
work_keys_str_mv AT vidaljoel brickognizeapplyingphotorealisticimagesynthesisforlegobricksrecognitionwithlimiteddata
AT vallicrosaguillem brickognizeapplyingphotorealisticimagesynthesisforlegobricksrecognitionwithlimiteddata
AT martirobert brickognizeapplyingphotorealisticimagesynthesisforlegobricksrecognitionwithlimiteddata
AT barnadamarc brickognizeapplyingphotorealisticimagesynthesisforlegobricksrecognitionwithlimiteddata