Cross-Stage Attention Propagation for Efficient Semantic Segmentation

2026-04-07 · Beoungwoo Kang arxiv

Recent lightweight semantic segmentation methods have made significant progress by combining compact backbones with efficient decoder heads. However, most multi-scale decoders compute attention independently at each feature scale, introducing substantial redundancy since the resulting attention distributions across scales are strongly correlated. We propose Cross-Stage Attention Propagation (CSAP), a decoder framework that computes attention at the deepest feature scale and propagates the resulting attention maps to shallower stages, bypassing query-key computation at those stages entirely. This design preserves multi-scale contextual reasoning while substantially reducing the decoder's computational cost. CSAP-Tiny achieves 42.9% mIoU on ADE20K with only 5.5 GFLOPs, 80.5% on Cityscapes with 21.5 GFLOPs, and 40.9% on COCO-Stuff 164K with 5.5 GFLOPs, surpassing SegNeXt-Tiny by +1.8% on ADE20K while requiring 16.8% fewer floating-point operations.

📄 PDF Abstract BibTeX arXiv:2604.05431

Code (0)

등록된 구현이 없습니다.

Tasks

Semantic Segmentation

Results from the Paper

Rank	Task	Dataset	Model	Metrics
#481	Semantic Segmentation	ADE20K	Cross-Stage	GFLOPs: 42.9
#3	Semantic Segmentation	COCO-Stuff	Cross-Stage	mIoU: 40.9
#5	Semantic Segmentation	Cityscapes	Cross-Stage	mIoU: 80.5

Biologically-inspired Semi-supervised Semantic Segmentation for Biomedical Imaging

2024-12-04 · Luca Ciampi, Gabriele Lagani, Giuseppe Amato, Fabrizio Falchi

We propose a novel two-stage semi-supervised learning approach for training downsampling-upsampling semantic segmentation architectures. The first stage does not use backpropagation. Rather, it exploits the bio-inspired …

Image Segmentation Medical Image Segmentation Semantic Segmentation Semi-Supervised Semantic Segmentation

GETAM: Gradient-weighted Element-wise Transformer Attention Map for Weakly-supervised Semantic segmentation

2021-12-06 · Weixuan Sun, Jing Zhang, Zheyuan Liu, Yiran Zhong 외

Weakly Supervised Semantic Segmentation (WSSS) is challenging, particularly when image-level labels are used to supervise pixel level prediction. To bridge their gap, a Class Activation Map (CAM) is usually generated to …

Semantic Segmentation Weakly supervised Semantic Segmentation Weakly-Supervised Semantic Segmentation

SSR: Semantic and Spatial Rectification for CLIP-based Weakly Supervised Segmentation

2025-12-01 · Xiuli Bi, Die Xiao, Junchao Fan, Bin Xiao arxiv

In recent years, Contrastive Language-Image Pretraining (CLIP) has been widely applied to Weakly Supervised Semantic Segmentation (WSSS) tasks due to its powerful cross-modal semantic understanding capabilities. This pap…

Semantic Segmentation Contrastive Learning

Global-Local Propagation Network for RGB-D Semantic Segmentation

2021-01-26 · Sihan Chen, Xinxin Zhu, Wei Liu, Xingjian He 외

Depth information matters in RGB-D semantic segmentation task for providing additional geometric information to color images. Most existing methods exploit a multi-stage fusion strategy to propagate depth feature to the …

Decoder Scene Segmentation Segmentation Semantic Segmentation

Semantics Meets Temporal Correspondence: Self-supervised Object-centric Learning in Videos

2023-08-19 · ICCV 2023 1 · Rui Qian, Shuangrui Ding, Xian Liu, Dahua Lin

Self-supervised methods have shown remarkable progress in learning high-level semantics and low-level temporal correspondence. Building on these results, we take one step further and explore the possibility of integratin…

Object Object Discovery Semantic Segmentation