EUVIMEDEUVIMED entdecken
Crossref

SPGrasp: Spatiotemporal Prompt-Driven Grasp Synthesis in Dynamic Scenes

Yunpeng Mei, Hongjie Cao, Wei Xiao, Yinqiu Xia, Zhaohan Feng, Gang Wang, Jie Chen

IEEE Journal of Selected Topics in Signal Processing · 2026-03 · 20 (2) · 202-212

DOI: 10.1109/jstsp.2026.3671182 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Journal of Selected Topics in Signal Processing, 2026-03. Verfasst von Yunpeng Mei und 6 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 41 Literaturangaben. Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Real-time interactive grasp synthesis for dynamic objects remains challenging, as existing instance-level methods struggle to achieve low-latency inference while maintaining robust temporal consistency. To bridge this gap, we propose SPGrasp (Spatiotemporal Prompt-driven dynamic Grasp synthesis), a novel framework that extends the Segment Anything Model 2 (SAM 2) for video-stream grasp estimation. Our core innovation integrates user prompts with a spatiotemporal context module, enabling real-time interaction with end-to-end latency as low as 59 ms while preserving consistent instance identity and grasp predictions in dynamic, cluttered scenes, including object overlap and occlusion. In benchmark evaluations, SPGrasp achieves instance-level grasp accuracies of 90.6% on OCID and 93.8% on Jacquard. On the GraspNet-1Billion dataset under continuous tracking, SPGrasp reaches 92.0% accuracy with 73.1 ms per-frame latency, corresponding to a 58.5% latency reduction over the prior state-of-the-art promptable method RoG-SAM while maintaining competitive accuracy. Real-world experiments further demonstrate reliable interactive grasping under frequent occlusions, achieving a 94.8% success rate. These results suggest that SPGrasp effectively mitigates the latency–interactivity trade-off in dynamic grasp synthesis.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Mei, Y., Cao, H., Xiao, W., Xia, Y., Feng, Z., Wang, G., & Chen, J. (2026). SPGrasp: Spatiotemporal Prompt-Driven Grasp Synthesis in Dynamic Scenes. IEEE Journal of Selected Topics in Signal Processing, 20(2), 202-212. https://doi.org/10.1109/jstsp.2026.3671182

Vancouver

Mei Y, Cao H, Xiao W, Xia Y, Feng Z, Wang G, et al. SPGrasp: Spatiotemporal Prompt-Driven Grasp Synthesis in Dynamic Scenes. IEEE Journal of Selected Topics in Signal Processing. 2026;20(2):202-212. doi:10.1109/jstsp.2026.3671182

BibTeX
@article{Mei2026,
  author = {Mei, Yunpeng and Cao, Hongjie and Xiao, Wei and Xia, Yinqiu and Feng, Zhaohan and Wang, Gang and Chen, Jie},
  title = {SPGrasp: Spatiotemporal Prompt-Driven Grasp Synthesis in Dynamic Scenes},
  journal = {IEEE Journal of Selected Topics in Signal Processing},
  year = {2026},
  volume = {20},
  number = {2},
  pages = {202--212},
  doi = {10.1109/jstsp.2026.3671182}
}
Kostenlos starten →