MTMamba++: Enhancing Multi-Task Dense Scene Understanding via Mamba-Based Decoders
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911077863784448 |
|---|---|
| author | Lin, Baijiong Jiang, Weisen Chen, Pengguang Liu, Shu Chen, Ying-Cong |
| author_facet | Lin, Baijiong Jiang, Weisen Chen, Pengguang Liu, Shu Chen, Ying-Cong |
| contents | Multi-task dense scene understanding, which trains a model for multiple dense prediction tasks, has a wide range of application scenarios. Capturing long-range dependency and enhancing cross-task interactions are crucial to multi-task dense prediction. In this paper, we propose MTMamba++, a novel architecture for multi-task scene understanding featuring with a Mamba-based decoder. It contains two types of core blocks: self-task Mamba (STM) block and cross-task Mamba (CTM) block. STM handles long-range dependency by leveraging state-space models, while CTM explicitly models task interactions to facilitate information exchange across tasks. We design two types of CTM block, namely F-CTM and S-CTM, to enhance cross-task interaction from feature and semantic perspectives, respectively. Extensive experiments on NYUDv2, PASCAL-Context, and Cityscapes datasets demonstrate the superior performance of MTMamba++ over CNN-based, Transformer-based, and diffusion-based methods while maintaining high computational efficiency. The code is available at https://github.com/EnVision-Research/MTMamba. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_15101 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MTMamba++: Enhancing Multi-Task Dense Scene Understanding via Mamba-Based Decoders Lin, Baijiong Jiang, Weisen Chen, Pengguang Liu, Shu Chen, Ying-Cong Computer Vision and Pattern Recognition Artificial Intelligence Multi-task dense scene understanding, which trains a model for multiple dense prediction tasks, has a wide range of application scenarios. Capturing long-range dependency and enhancing cross-task interactions are crucial to multi-task dense prediction. In this paper, we propose MTMamba++, a novel architecture for multi-task scene understanding featuring with a Mamba-based decoder. It contains two types of core blocks: self-task Mamba (STM) block and cross-task Mamba (CTM) block. STM handles long-range dependency by leveraging state-space models, while CTM explicitly models task interactions to facilitate information exchange across tasks. We design two types of CTM block, namely F-CTM and S-CTM, to enhance cross-task interaction from feature and semantic perspectives, respectively. Extensive experiments on NYUDv2, PASCAL-Context, and Cityscapes datasets demonstrate the superior performance of MTMamba++ over CNN-based, Transformer-based, and diffusion-based methods while maintaining high computational efficiency. The code is available at https://github.com/EnVision-Research/MTMamba. |
| title | MTMamba++: Enhancing Multi-Task Dense Scene Understanding via Mamba-Based Decoders |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2408.15101 |