Occluded target grasping detection method based on spatial information aggregation
Information Sciences|更新时间:2024-10-28
|
Occluded target grasping detection method based on spatial information aggregation
“In the field of robot vision grasping, researchers have proposed a pose detection method for occluded targets based on spatial information aggregation, which effectively improves the detection accuracy of occluded targets. This method enhances the network's understanding of target structures and shapes through coordinate convolution and spatial information aggregation modules. Experimental verification on multiple datasets shows that this method has the highest detection accuracy and performs better in practical scenarios.”
Optics and Precision EngineeringVol. 32, Issue 18, Pages: 2792-2802(2024)
CHEN Renxiang,QIU Tianran,YANG Lixia,et al.Occluded target grasping detection method based on spatial information aggregation[J].Optics and Precision Engineering,2024,32(18):2792-2802.
CHEN Renxiang,QIU Tianran,YANG Lixia,et al.Occluded target grasping detection method based on spatial information aggregation[J].Optics and Precision Engineering,2024,32(18):2792-2802. DOI: 10.37188/OPE.20243218.2792.
Occluded target grasping detection method based on spatial information aggregation