MiPa: Mixed Patch Infrared-Visible Modality Agnostic Object Detection

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Medeiros, Heitor R., Latortue, David, Granger, Eric, Pedersoli, Marco
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914896906551296
author Medeiros, Heitor R.
Latortue, David
Granger, Eric
Pedersoli, Marco
author_facet Medeiros, Heitor R.
Latortue, David
Granger, Eric
Pedersoli, Marco
contents In real-world scenarios, using multiple modalities like visible (RGB) and infrared (IR) can greatly improve the performance of a predictive task such as object detection (OD). Multimodal learning is a common way to leverage these modalities, where multiple modality-specific encoders and a fusion module are used to improve performance. In this paper, we tackle a different way to employ RGB and IR modalities, where only one modality or the other is observed by a single shared vision encoder. This realistic setting requires a lower memory footprint and is more suitable for applications such as autonomous driving and surveillance, which commonly rely on RGB and IR data. However, when learning a single encoder on multiple modalities, one modality can dominate the other, producing uneven recognition results. This work investigates how to efficiently leverage RGB and IR modalities to train a common transformer-based OD vision encoder, while countering the effects of modality imbalance. For this, we introduce a novel training technique to Mix Patches (MiPa) from the two modalities, in conjunction with a patch-wise modality agnostic module, for learning a common representation of both modalities. Our experiments show that MiPa can learn a representation to reach competitive results on traditional RGB/IR benchmarks while only requiring a single modality during inference. Our code is available at: https://github.com/heitorrapela/MiPa.
format Preprint
id arxiv_https___arxiv_org_abs_2404_18849
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MiPa: Mixed Patch Infrared-Visible Modality Agnostic Object Detection
Medeiros, Heitor R.
Latortue, David
Granger, Eric
Pedersoli, Marco
Computer Vision and Pattern Recognition
In real-world scenarios, using multiple modalities like visible (RGB) and infrared (IR) can greatly improve the performance of a predictive task such as object detection (OD). Multimodal learning is a common way to leverage these modalities, where multiple modality-specific encoders and a fusion module are used to improve performance. In this paper, we tackle a different way to employ RGB and IR modalities, where only one modality or the other is observed by a single shared vision encoder. This realistic setting requires a lower memory footprint and is more suitable for applications such as autonomous driving and surveillance, which commonly rely on RGB and IR data. However, when learning a single encoder on multiple modalities, one modality can dominate the other, producing uneven recognition results. This work investigates how to efficiently leverage RGB and IR modalities to train a common transformer-based OD vision encoder, while countering the effects of modality imbalance. For this, we introduce a novel training technique to Mix Patches (MiPa) from the two modalities, in conjunction with a patch-wise modality agnostic module, for learning a common representation of both modalities. Our experiments show that MiPa can learn a representation to reach competitive results on traditional RGB/IR benchmarks while only requiring a single modality during inference. Our code is available at: https://github.com/heitorrapela/MiPa.
title MiPa: Mixed Patch Infrared-Visible Modality Agnostic Object Detection
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2404.18849