{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/multimodal-transformer-for-material","title":"MMSFormer: Multimodal Transformer for Material and Semantic Segmentation","arxiv_id":"2309.04001","date":"2023-09-07","proceeding":null,"authors":["Md Kaykobad Reza","Ashley Prater-Bennette","M. Salman Asif"],"abstract":"Leveraging information across diverse modalities is known to enhance performance on multimodal segmentation tasks. However, effectively fusing information from different modalities remains challenging due to the unique characteristics of each modality. In this paper, we propose a novel fusion strategy that can effectively fuse information from different modality combinations. We also propose a new model named Multi-Modal Segmentation TransFormer (MMSFormer) that incorporates the proposed fusion strategy to perform multimodal material and semantic segmentation tasks. MMSFormer outperforms current state-of-the-art models on three different datasets. As we begin with only one input modality, performance improves progressively as additional modalities are incorporated, showcasing the effectiveness of the fusion block in combining useful information from diverse input modalities. Ablation studies show that different modules in the fusion block are crucial for overall model performance. Furthermore, our ablation studies also highlight the capacity of different input modalities to improve performance in the identification of different types of materials. The code and pretrained models will be made available at https://github.com/csiplab/MMSFormer.","url_abs":"https://arxiv.org/abs/2309.04001v4","url_pdf":"https://arxiv.org/pdf/2309.04001v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"multimodal-transformer-for-material","repo_url":"https://github.com/csiplab/mmsformer","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"segmentation","task_name":"Segmentation"},{"task_slug":"semantic-segmentation","task_name":"Semantic Segmentation"},{"task_slug":"thermal-image-segmentation","task_name":"Thermal Image Segmentation"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semantic-segmentation-on-fmb-dataset","task":"Semantic Segmentation","dataset":"FMB Dataset","model":"MMSFormer (RGB-Infrared)","rank_in_archive_order":5,"of":14,"metrics":{"mIoU":"61.70"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-fmb-dataset","task":"Semantic Segmentation","dataset":"FMB Dataset","model":"MMSFormer (RGB)","rank_in_archive_order":6,"of":14,"metrics":{"mIoU":"57.20"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-mcubes","task":"Semantic Segmentation","dataset":"MCubeS","model":"MMSFormer (RGB-A-D-N)","rank_in_archive_order":4,"of":22,"metrics":{"mIoU":"53.11%"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-mcubes","task":"Semantic Segmentation","dataset":"MCubeS","model":"MMSFormer (RGB-A-D)","rank_in_archive_order":9,"of":22,"metrics":{"mIoU":"52.05%"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-mcubes","task":"Semantic Segmentation","dataset":"MCubeS","model":"MMSFormer (RGB-A)","rank_in_archive_order":11,"of":22,"metrics":{"mIoU":"51.30%"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-mcubes","task":"Semantic Segmentation","dataset":"MCubeS","model":"MMSFormer (RGB)","rank_in_archive_order":15,"of":22,"metrics":{"mIoU":"50.44%"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-mcubes-p","task":"Semantic Segmentation","dataset":"MCubeS (P)","model":"MMSFormer (RGB-A-D)","rank_in_archive_order":1,"of":8,"metrics":{"mIoU":"52.03"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-mcubes-p","task":"Semantic Segmentation","dataset":"MCubeS (P)","model":"MMSFormer (RGB-A)","rank_in_archive_order":2,"of":8,"metrics":{"mIoU":"51.30"},"uses_additional_data":false},{"leaderboard":"/sota/semantic-segmentation-on-mcubes-p","task":"Semantic Segmentation","dataset":"MCubeS (P)","model":"MMSFormer (RGB)","rank_in_archive_order":5,"of":8,"metrics":{"mIoU":"50.44"},"uses_additional_data":false},{"leaderboard":"/sota/thermal-image-segmentation-on-pst900","task":"Thermal Image Segmentation","dataset":"PST900","model":"MMSFormer","rank_in_archive_order":5,"of":22,"metrics":{"mIoU":"87.45"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}