Shuobin Wei
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Shuobin Wei has a total of 4 co-authors. Here are the top co-authors by number of shared publications:
- Binghua Su
Beijing Institute of Technology, Zhuhai 519088, China
1 shared publication - Zhengan Lu
School of Information Technology, Beijing Institute of Technology, Zhuhai, China
1 shared publication - Zhuang Zhou
School of Information and Electronics, Beijing Institute of Technology, Zhuhai, China
1 shared publication - Zizhao Yuan
Beijing Institute of Technology, Zhuhai, China
1 shared publication
Latest Publications
- 12 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 32 (10709908) , 91-95
HDBFormer: Efficient RGB-D Semantic Segmentation With a Heterogeneous Dual-Branch Framework
In RGB-D semantic segmentation for indoor scenes, a key challenge is effectively integrating the rich color information from RGB images with the spatial distance information from depth images. However, most existing methods overlook the inherent differences in how RGB and depth images express information. Properly distinguishing the processing of RGB and depth images is essential to fully exploiting their unique and significant characteristics. To address this, we propose a novel heterogeneous dual-branch framework called HDBFormer, specifically designed to handle these modality differences. For RGB images, which contain rich detail, we employ both a basic and detail encoder to extract local and global features. For the simpler depth images, we propose LDFormer, a lightweight hierarchical encoder that efficiently extracts depth features with fewer parameters. Additionally, we introduce the Modality Information Interaction Module (MIIM), which combines transformers with large kernel convolutions to interact global and local information across modalities efficiently. Extensive experiments show that HDBFormer achieves state-of-the-art performance on the NYUDepthv2 and SUN-RGBD datasets.
- 12 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Signal Processing Letters
- Vol. 32 (10709908) , 91-95
HDBFormer: Efficient RGB-D Semantic Segmentation With a Heterogeneous Dual-Branch Framework
In RGB-D semantic segmentation for indoor scenes, a key challenge is effectively integrating the rich color information from RGB images with the spatial distance information from depth images. However, most existing methods overlook the inherent differences in how RGB and depth images express information. Properly distinguishing the processing of RGB and depth images is essential to fully exploiting their unique and significant characteristics. To address this, we propose a novel heterogeneous dual-branch framework called HDBFormer, specifically designed to handle these modality differences. For RGB images, which contain rich detail, we employ both a basic and detail encoder to extract local and global features. For the simpler depth images, we propose LDFormer, a lightweight hierarchical encoder that efficiently extracts depth features with fewer parameters. Additionally, we introduce the Modality Information Interaction Module (MIIM), which combines transformers with large kernel convolutions to interact global and local information across modalities efficiently. Extensive experiments show that HDBFormer achieves state-of-the-art performance on the NYUDepthv2 and SUN-RGBD datasets.