.png){kind=icon}
.png){kind=icon}
Return
Global feature enhancement and skip-connected fusion for grasping detection
.png){kind=icon}
PRE
AI
S
X
R
Y
B
E
X
DOI：10.1007/s10514-026-10242-x.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
In unstructured environments, grasping tasks face challenges in grasping pose estimation and insufficient real-time performance. Existing methods struggle to suppress background noise and highlight critical grasping regions, leading to false detections or positioning errors in cluttered environments. To address this, we propose the Grasp Detection Network GFESF-Grasp, which integrates Global Feature Enhancement (GFE) and Skip-Connected Feature Fusion (SF). The GFE module is designed at the bottleneck layer, combining parallel dilated convolutions with differential dilation rates to capture both local details and global context. This expands the effective receptive field, enhancing the network’s ability to perceive cross-scale geometric features of objects. It also integrates Mamba-Like Linear Attention (MLLA) to strengthen feature responses in critical regions like object edges, improving the network’s discriminative power under noisy conditions. During decoding, the SF module adaptively fuses cross-level features, effectively suppressing redundant information while preserving key features. The proposed network achieves 98.9%, 96.1% and 87.0% accuracy on Cornell, Jacquard and OCID-Grasp grasping datasets respectively. In complex scenarios, the grasping success rate for unknown objects reached 92.2%, validating its effectiveness and practicality in challenging environments. Code are available at: https://github.com/wangxauat/gfesf-grasp.
Keywords:
Global feature enhancement
Multi-scale feature fusion
Attention mechanism
Unstructured environment
Robotic grasping detection
Journal
IF:
4.3
Papers:
1.6K
Citations:
5.0K
Organization
Cited Papers