GaMi: Geometry-Agnostic Material Identification via Cross-Modal Subtractive Disentanglement
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917547137302528 |
|---|---|
| author | Chen, Zhiwei Li, Yijie Zhang, Yimo Shao, Shiyun Chen, Yichao Ding, Dian Wang, Liang Wu, Haiwei Guo, Liwei Yang, Jie Zhang, Xiaosong Zhang, Yongzhao |
| author_facet | Chen, Zhiwei Li, Yijie Zhang, Yimo Shao, Shiyun Chen, Yichao Ding, Dian Wang, Liang Wu, Haiwei Guo, Liwei Yang, Jie Zhang, Xiaosong Zhang, Yongzhao |
| contents | Non-contact material identification enables adaptive interaction for embodied intelligence yet faces challenges from geometry-induced variations (e.g., orientation, shape, distance) and single-modality ambiguities. In this paper, we present GaMi, a multimodal material identification system integrating mmWave and acoustic sensing to robustly operate under unconstrained geometric conditions. By leveraging the insight of shared geometric consistency between co-located bimodal sensors, GaMi employs an intra-sample cross-modal subtractive disentanglement framework. By semantically aligning modalities and subtracting the shared geometric context, it isolates intrinsic material features. Furthermore, GaMi incorporates inter-sample contrastive learning to correct the residual interference caused by cross-modal misalignment. Additionally, a pairing-based adaptation strategy between two modalities enables few-shot generalization across devices. Extensive evaluations on 20 materials show that GaMi achieves 95.2% accuracy, outperforming single-modality baselines across unseen geometric conditions. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_30818 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | GaMi: Geometry-Agnostic Material Identification via Cross-Modal Subtractive Disentanglement Chen, Zhiwei Li, Yijie Zhang, Yimo Shao, Shiyun Chen, Yichao Ding, Dian Wang, Liang Wu, Haiwei Guo, Liwei Yang, Jie Zhang, Xiaosong Zhang, Yongzhao Emerging Technologies Artificial Intelligence Sound Non-contact material identification enables adaptive interaction for embodied intelligence yet faces challenges from geometry-induced variations (e.g., orientation, shape, distance) and single-modality ambiguities. In this paper, we present GaMi, a multimodal material identification system integrating mmWave and acoustic sensing to robustly operate under unconstrained geometric conditions. By leveraging the insight of shared geometric consistency between co-located bimodal sensors, GaMi employs an intra-sample cross-modal subtractive disentanglement framework. By semantically aligning modalities and subtracting the shared geometric context, it isolates intrinsic material features. Furthermore, GaMi incorporates inter-sample contrastive learning to correct the residual interference caused by cross-modal misalignment. Additionally, a pairing-based adaptation strategy between two modalities enables few-shot generalization across devices. Extensive evaluations on 20 materials show that GaMi achieves 95.2% accuracy, outperforming single-modality baselines across unseen geometric conditions. |
| title | GaMi: Geometry-Agnostic Material Identification via Cross-Modal Subtractive Disentanglement |
| topic | Emerging Technologies Artificial Intelligence Sound |
| url | https://arxiv.org/abs/2605.30818 |