EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Boundary-Aware Multi-Task Knowledge Distillation for Lightweight Grasp Detection

Lei Qiao, Dongshi Bian, Wugui Wang, Bing Li, Bowen Xing

Electronics · 18.11.2026 · 15 (22) · 5379

DOI: 10.3390/electronics15225379 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 18.11.2026. Verfasst von Lei Qiao und 4 weiteren Autor:innen. Die Arbeit nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Lightweight grasp detectors suit resource-constrained robotic platforms, but compact encoders lose the spatial discrimination needed to keep grasp quality, orientation, and opening width consistent at one location, and generic distillation transfers teacher outputs without testing whether their grasp geometry is valid. This paper presents BA-MTKD, a boundary-aware multi-task distillation framework that transfers only teacher outputs whose quality, angle, and width agree with the training annotation and couples quality transitions with local angle–width relations at grasp boundaries. All teacher-dependent components are discarded after training. Against sixteen distillation methods retrained under one matched RGB protocol, a MobileNetV3-Small student trained with BA-MTKD achieves the highest accuracy on Cornell, OCID-Grasp, and VMRD V2 (97.6%, 84.7%, and 77.9%) and recovers 84–90% of the teacher–student gap. Its advantage widens under the strict criterion (IoU 0.50, 15°): it reaches 89.1%, 68.4%, and 59.8%, exceeds the strongest competitor by 2.0–3.3 points, and comes within 0.9 points of a twelve-times-larger ResNet-18 teacher. BA-MTKD further reduces the dense angle error by 26–29% and shows the smallest degradation under four image perturbations, and all six primary gains are significant after Holm adjustment. The deployed student has 1.00 M parameters and 0.179 G MACs and runs in 3.4 ms per image on a desktop GPU. These offline benchmark results establish BA-MTKD as a compact perception module for the next validation stage: closed-loop robotic grasping on embedded hardware.

Zitieren

APA

Qiao, L., Bian, D., Wang, W., Li, B., & Xing, B. (2026). Boundary-Aware Multi-Task Knowledge Distillation for Lightweight Grasp Detection. Electronics, 15(22), 5379. https://doi.org/10.3390/electronics15225379

Vancouver

Qiao L, Bian D, Wang W, Li B, Xing B. Boundary-Aware Multi-Task Knowledge Distillation for Lightweight Grasp Detection. Electronics. 2026;15(22):5379. doi:10.3390/electronics15225379

BibTeX
@article{Qiao2026,
  author = {Qiao, Lei and Bian, Dongshi and Wang, Wugui and Li, Bing and Xing, Bowen},
  title = {Boundary-Aware Multi-Task Knowledge Distillation for Lightweight Grasp Detection},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {22},
  pages = {5379},
  doi = {10.3390/electronics15225379}
}
Kostenlos starten →