EUVIMEDEUVIMED entdecken
Crossref

Prototype-Guided Routed Analytic Learning for Exemplar-Free Incremental Human-Object Interaction Detection

Kunyuan Deng, Yi Wang, Huiping Zhuang, Yawen Cui, Ying Zheng, Lap-Pui Chau

IEEE Transactions on Circuits and Systems for Video Technology · 2026-11 · 36 (11) · 16206-16221

DOI: 10.1109/tcsvt.2026.3711166 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2026-11. Verfasst von Kunyuan Deng und 5 weiteren Autor:innen. Die Arbeit nennt 65 Literaturangaben. Gefördert von: JC STEM Laboratory of Machine Learning and Computer Vision funded by The Hong Kong Jockey Club Charities Trust, Global STEM Professorship Scheme from Hong Kong Special Administrative Region, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Human-object interaction (HOI) detection is crucial for advanced scene understanding, but real-world deployment requires models that continually acquire new interactions without forgetting previous knowledge. This incremental setting is especially challenging in HOI due to two coupled issues: catastrophic forgetting and severe multi-label long-tailed imbalance, particularly under privacy-sensitive applications where revisiting past data is infeasible. In this paper, we introduce Prototype-Guided Routed Analytic Learning (PGRAL), an exemplar-free framework that jointly addresses both challenges. PGRAL first establishes a highly discriminative representation space via prototype-guided contrastive refinement for pre-training, enhancing intra-class compactness and inter-class separability. For incremental learning, our routed analytic incremental learner recursively updates the classifier in closed form, effectively mitigating task-recency bias and catastrophic forgetting. To address the long-tailed imbalance, we further devise a knowledge-enhanced feature synthesis module that generates high-quality features for tail classes, with the learner’s routing mechanism enabling stable and effective integration. Extensive experiments on HICO-DET and V-COCO demonstrate that PGRAL achieves new state-of-the-art performance, with particularly significant gains on rare tail interactions, offering an effective and scalable solution for HOI understanding in dynamic, real-world environments.

Förderung

  • JC STEM Laboratory of Machine Learning and Computer Vision funded by The Hong Kong Jockey Club Charities Trust
  • Global STEM Professorship Scheme from Hong Kong Special Administrative Region
  • National Natural Science Foundation of China
  • Guangdong S&T Programme
  • Guangdong Basic and Applied Basic Research Foundation
  • GJYC program of Guangzhou

Zitieren

APA

Deng, K., Wang, Y., Zhuang, H., Cui, Y., Zheng, Y., & Chau, L. (2026). Prototype-Guided Routed Analytic Learning for Exemplar-Free Incremental Human-Object Interaction Detection. IEEE Transactions on Circuits and Systems for Video Technology, 36(11), 16206-16221. https://doi.org/10.1109/tcsvt.2026.3711166

Vancouver

Deng K, Wang Y, Zhuang H, Cui Y, Zheng Y, Chau L. Prototype-Guided Routed Analytic Learning for Exemplar-Free Incremental Human-Object Interaction Detection. IEEE Transactions on Circuits and Systems for Video Technology. 2026;36(11):16206-16221. doi:10.1109/tcsvt.2026.3711166

BibTeX
@article{Deng2026,
  author = {Deng, Kunyuan and Wang, Yi and Zhuang, Huiping and Cui, Yawen and Zheng, Ying and Chau, Lap-Pui},
  title = {Prototype-Guided Routed Analytic Learning for Exemplar-Free Incremental Human-Object Interaction Detection},
  journal = {IEEE Transactions on Circuits and Systems for Video Technology},
  year = {2026},
  volume = {36},
  number = {11},
  pages = {16206--16221},
  doi = {10.1109/tcsvt.2026.3711166}
}
Kostenlos starten →