EUVIMEDEUVIMED entdecken
Crossref

VLPrompt-PSG: Vision-Language Prompting for Panoptic Scene Graph Generation

Zijian Zhou, Holger Caesar, Qijun Chen, Miaojing Shi

International Journal of Computer Vision · 2025-11 · 133 (11) · 8006-8021

DOI: 10.1007/s11263-025-02564-7 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Computer Vision, online erschienen am 23.08.2025. Verfasst von Zijian Zhou und 3 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 98 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Panoptic scene graph generation (PSG) aims at achieving a comprehensive image understanding by simultaneously segmenting objects and predicting relations among objects. However, the long-tail problem among relations leads to unsatisfactory results in real-world applications. Prior methods predominantly rely on vision information or utilize limited language information, such as object or relation names, thereby overlooking the utility of language information. Leveraging the recent progress in Large Language Models (LLMs), we propose to use language information to assist relation prediction, particularly for rare relations. To this end, we propose the Vision-Language Prompting (VLPrompt) model, which acquires vision information from images and language information from LLMs. Then, through a prompter network based on attention mechanism, it achieves precise relation prediction. Our extensive experiments show that VLPrompt significantly outperforms previous state-of-the-art methods on the PSG dataset, proving the effectiveness of incorporating language information and alleviating the long-tail problem of relations. Code is available at https://github.com/franciszzj/VLPrompt.

Zitieren

APA

Zhou, Z., Caesar, H., Chen, Q., & Shi, M. (2025). VLPrompt-PSG: Vision-Language Prompting for Panoptic Scene Graph Generation. International Journal of Computer Vision, 133(11), 8006-8021. https://doi.org/10.1007/s11263-025-02564-7

Vancouver

Zhou Z, Caesar H, Chen Q, Shi M. VLPrompt-PSG: Vision-Language Prompting for Panoptic Scene Graph Generation. International Journal of Computer Vision. 2025;133(11):8006-8021. doi:10.1007/s11263-025-02564-7

BibTeX
@article{Zhou2025,
  author = {Zhou, Zijian and Caesar, Holger and Chen, Qijun and Shi, Miaojing},
  title = {VLPrompt-PSG: Vision-Language Prompting for Panoptic Scene Graph Generation},
  journal = {International Journal of Computer Vision},
  year = {2025},
  volume = {133},
  number = {11},
  pages = {8006--8021},
  doi = {10.1007/s11263-025-02564-7}
}
Kostenlos starten →