RGB-to-Infrared Translation Using Ensemble Learning Applied to Driving Scenarios

Multimodal sensing is essential in order to reach the robustness required of autonomous vehicle perception systems. Infrared (IR) imaging is of particular interest due to its low cost and complementarity with traditional RGB sensors. However, the lack of IR data in many datasets and simulation tools...

Full description

Saved in:
Bibliographic Details
Main Authors: Leonardo Ravaglia, Roberto Longo, Kaili Wang, David Van Hamme, Julie Moeyersoms, Ben Stoffelen, Tom De Schepper
Format: Article
Language:English
Published: MDPI AG 2025-06-01
Series:Journal of Imaging
Subjects:
Online Access:https://www.mdpi.com/2313-433X/11/7/206
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850078030980972544
author Leonardo Ravaglia
Roberto Longo
Kaili Wang
David Van Hamme
Julie Moeyersoms
Ben Stoffelen
Tom De Schepper
author_facet Leonardo Ravaglia
Roberto Longo
Kaili Wang
David Van Hamme
Julie Moeyersoms
Ben Stoffelen
Tom De Schepper
author_sort Leonardo Ravaglia
collection DOAJ
description Multimodal sensing is essential in order to reach the robustness required of autonomous vehicle perception systems. Infrared (IR) imaging is of particular interest due to its low cost and complementarity with traditional RGB sensors. However, the lack of IR data in many datasets and simulation tools limits the development and validation of sensor fusion algorithms that exploit this complementarity. To address this, we propose an augmentation method that synthesizes realistic IR data from RGB images using gradient-boosting decision trees. We demonstrate that this method is an effective alternative to traditional deep learning methods for image translation such as CNNs and GANs, particularly in data-scarce situations. The proposed approach generates high-quality synthetic IR, i.e., Near-Infrared (NIR) and thermal images from RGB images, enhancing datasets such as MS2, EPFL, and Freiburg. Our synthetic images exhibit good visual quality when evaluated using metrics such as <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>R</mi><mn>2</mn></msup></semantics></math></inline-formula>, PSNR, SSIM, and LPIPS, achieving an <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>R</mi><mn>2</mn></msup></semantics></math></inline-formula> of 0.98 on the MS2 dataset and a PSNR of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>21.3</mn></mrow></semantics></math></inline-formula> dB on the Freiburg dataset. We also discuss the application of this method to synthetic RGB images generated by the CARLA simulator for autonomous driving. Our approach provides richer datasets with a particular focus on IR modalities for sensor fusion along with a framework for generating a wider variety of driving scenarios within urban driving datasets, which can help to enhance the robustness of sensor fusion algorithms.
format Article
id doaj-art-511870f7c62b4ad083f9e581ab5a19c4
institution DOAJ
issn 2313-433X
language English
publishDate 2025-06-01
publisher MDPI AG
record_format Article
series Journal of Imaging
spelling doaj-art-511870f7c62b4ad083f9e581ab5a19c42025-08-20T02:45:39ZengMDPI AGJournal of Imaging2313-433X2025-06-0111720610.3390/jimaging11070206RGB-to-Infrared Translation Using Ensemble Learning Applied to Driving ScenariosLeonardo Ravaglia0Roberto Longo1Kaili Wang2David Van Hamme3Julie Moeyersoms4Ben Stoffelen5Tom De Schepper6Interuniversity Microelectronics Centre, Kapeldreef 75, 3001 Leuven, BelgiumInteruniversity Microelectronics Centre, Kapeldreef 75, 3001 Leuven, BelgiumInteruniversity Microelectronics Centre, Kapeldreef 75, 3001 Leuven, BelgiumInteruniversity Microelectronics Centre, Kapeldreef 75, 3001 Leuven, BelgiumInteruniversity Microelectronics Centre, Kapeldreef 75, 3001 Leuven, BelgiumInteruniversity Microelectronics Centre, Kapeldreef 75, 3001 Leuven, BelgiumInteruniversity Microelectronics Centre, Kapeldreef 75, 3001 Leuven, BelgiumMultimodal sensing is essential in order to reach the robustness required of autonomous vehicle perception systems. Infrared (IR) imaging is of particular interest due to its low cost and complementarity with traditional RGB sensors. However, the lack of IR data in many datasets and simulation tools limits the development and validation of sensor fusion algorithms that exploit this complementarity. To address this, we propose an augmentation method that synthesizes realistic IR data from RGB images using gradient-boosting decision trees. We demonstrate that this method is an effective alternative to traditional deep learning methods for image translation such as CNNs and GANs, particularly in data-scarce situations. The proposed approach generates high-quality synthetic IR, i.e., Near-Infrared (NIR) and thermal images from RGB images, enhancing datasets such as MS2, EPFL, and Freiburg. Our synthetic images exhibit good visual quality when evaluated using metrics such as <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>R</mi><mn>2</mn></msup></semantics></math></inline-formula>, PSNR, SSIM, and LPIPS, achieving an <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><msup><mi>R</mi><mn>2</mn></msup></semantics></math></inline-formula> of 0.98 on the MS2 dataset and a PSNR of <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>21.3</mn></mrow></semantics></math></inline-formula> dB on the Freiburg dataset. We also discuss the application of this method to synthetic RGB images generated by the CARLA simulator for autonomous driving. Our approach provides richer datasets with a particular focus on IR modalities for sensor fusion along with a framework for generating a wider variety of driving scenarios within urban driving datasets, which can help to enhance the robustness of sensor fusion algorithms.https://www.mdpi.com/2313-433X/11/7/206machine learningimage processingdata augmentationautonomous driving
spellingShingle Leonardo Ravaglia
Roberto Longo
Kaili Wang
David Van Hamme
Julie Moeyersoms
Ben Stoffelen
Tom De Schepper
RGB-to-Infrared Translation Using Ensemble Learning Applied to Driving Scenarios
Journal of Imaging
machine learning
image processing
data augmentation
autonomous driving
title RGB-to-Infrared Translation Using Ensemble Learning Applied to Driving Scenarios
title_full RGB-to-Infrared Translation Using Ensemble Learning Applied to Driving Scenarios
title_fullStr RGB-to-Infrared Translation Using Ensemble Learning Applied to Driving Scenarios
title_full_unstemmed RGB-to-Infrared Translation Using Ensemble Learning Applied to Driving Scenarios
title_short RGB-to-Infrared Translation Using Ensemble Learning Applied to Driving Scenarios
title_sort rgb to infrared translation using ensemble learning applied to driving scenarios
topic machine learning
image processing
data augmentation
autonomous driving
url https://www.mdpi.com/2313-433X/11/7/206
work_keys_str_mv AT leonardoravaglia rgbtoinfraredtranslationusingensemblelearningappliedtodrivingscenarios
AT robertolongo rgbtoinfraredtranslationusingensemblelearningappliedtodrivingscenarios
AT kailiwang rgbtoinfraredtranslationusingensemblelearningappliedtodrivingscenarios
AT davidvanhamme rgbtoinfraredtranslationusingensemblelearningappliedtodrivingscenarios
AT juliemoeyersoms rgbtoinfraredtranslationusingensemblelearningappliedtodrivingscenarios
AT benstoffelen rgbtoinfraredtranslationusingensemblelearningappliedtodrivingscenarios
AT tomdeschepper rgbtoinfraredtranslationusingensemblelearningappliedtodrivingscenarios