Enregistré dans:
Détails bibliographiques
Auteurs principaux: Lyu, Kailin, Xiao, Long, Zeng, Jianing, Dong, Junhao, Liu, Xuexin, Zou, Zhuojun, Yang, Haoyue, Shu, Lin, Hao, Jie
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2511.19509
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866909922314158080
author Lyu, Kailin
Xiao, Long
Zeng, Jianing
Dong, Junhao
Liu, Xuexin
Zou, Zhuojun
Yang, Haoyue
Shu, Lin
Hao, Jie
author_facet Lyu, Kailin
Xiao, Long
Zeng, Jianing
Dong, Junhao
Liu, Xuexin
Zou, Zhuojun
Yang, Haoyue
Shu, Lin
Hao, Jie
contents Traditional vision-based material perception methods often experience substantial performance degradation under visually impaired conditions, thereby motivating the shift toward non-visual multimodal material perception. Despite this, existing approaches frequently perform naive fusion of multimodal inputs, overlooking key challenges such as modality-specific noise, missing modalities common in real-world scenarios, and the dynamically varying importance of each modality depending on the task. These limitations lead to suboptimal performance across several benchmark tasks. In this paper, we propose a robust multimodal fusion framework, TouchFormer. Specifically, we employ a Modality-Adaptive Gating (MAG) mechanism and intra- and inter-modality attention mechanisms to adaptively integrate cross-modal features, enhancing model robustness. Additionally, we introduce a Cross-Instance Embedding Regularization(CER) strategy, which significantly improves classification accuracy in fine-grained subcategory material recognition tasks. Experimental results demonstrate that, compared to existing non-visual methods, the proposed TouchFormer framework achieves classification accuracy improvements of 2.48% and 6.83% on SSMC and USMC tasks, respectively. Furthermore, real-world robotic experiments validate TouchFormer's effectiveness in enabling robots to better perceive and interpret their environment, paving the way for its deployment in safety-critical applications such as emergency response and industrial automation. The code and datasets will be open-source, and the videos are available in the supplementary materials.
format Preprint
id arxiv_https___arxiv_org_abs_2511_19509
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TouchFormer: A Robust Transformer-based Framework for Multimodal Material Perception
Lyu, Kailin
Xiao, Long
Zeng, Jianing
Dong, Junhao
Liu, Xuexin
Zou, Zhuojun
Yang, Haoyue
Shu, Lin
Hao, Jie
Machine Learning
Traditional vision-based material perception methods often experience substantial performance degradation under visually impaired conditions, thereby motivating the shift toward non-visual multimodal material perception. Despite this, existing approaches frequently perform naive fusion of multimodal inputs, overlooking key challenges such as modality-specific noise, missing modalities common in real-world scenarios, and the dynamically varying importance of each modality depending on the task. These limitations lead to suboptimal performance across several benchmark tasks. In this paper, we propose a robust multimodal fusion framework, TouchFormer. Specifically, we employ a Modality-Adaptive Gating (MAG) mechanism and intra- and inter-modality attention mechanisms to adaptively integrate cross-modal features, enhancing model robustness. Additionally, we introduce a Cross-Instance Embedding Regularization(CER) strategy, which significantly improves classification accuracy in fine-grained subcategory material recognition tasks. Experimental results demonstrate that, compared to existing non-visual methods, the proposed TouchFormer framework achieves classification accuracy improvements of 2.48% and 6.83% on SSMC and USMC tasks, respectively. Furthermore, real-world robotic experiments validate TouchFormer's effectiveness in enabling robots to better perceive and interpret their environment, paving the way for its deployment in safety-critical applications such as emergency response and industrial automation. The code and datasets will be open-source, and the videos are available in the supplementary materials.
title TouchFormer: A Robust Transformer-based Framework for Multimodal Material Perception
topic Machine Learning
url https://arxiv.org/abs/2511.19509