EUVIMEDEUVIMED entdecken
Crossref

Decoupling What to Count and Where to See for Referring Expression Counting

Yuda Zou, Zijian Zhang, Yongchao Xu

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (16) · 14113-14121

DOI: 10.1609/aaai.v40i16.38423 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Yuda Zou und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Referring Expression Counting (REC) extends class-level object counting to the fine-grained subclass-level, aiming to enumerate objects matching a textual expression that specifies both the class and distinguishing attribute. A fundamental challenge, however, has been overlooked: annotation points are typically placed on class-representative locations (e.g., heads), forcing models to focus on class-level features while neglecting attribute information from other visual regions (e.g., legs for ''walking''). To address this, we propose W2-Net, a novel framework that explicitly decouples the problem into ''what to count'' and ''where to see'' via a dual-query mechanism. Specifically, alongside the standard what-to-count (w2c) queries that localize the object, we introduce dedicated where-to-see (w2s) queries. The w2s queries are guided to seek and extract features from attribute-specific visual regions, enabling precise subclass discrimination. Furthermore, we introduce Subclass Separable Matching (SSM), a novel matching strategy that incorporates a repulsive force to enhance inter-subclass separability during label assignment. W2-Net significantly outperforms the state-of-the-art on the REC-8K dataset, reducing counting error by 22.5% (validation) and 18.0% (test), and improving localization F1 by 7% and 8%, respectively.

Zitieren

APA

Zou, Y., Zhang, Z., & Xu, Y. (2026). Decoupling What to Count and Where to See for Referring Expression Counting. Proceedings of the AAAI Conference on Artificial Intelligence, 40(16), 14113-14121. https://doi.org/10.1609/aaai.v40i16.38423

Vancouver

Zou Y, Zhang Z, Xu Y. Decoupling What to Count and Where to See for Referring Expression Counting. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(16):14113-14121. doi:10.1609/aaai.v40i16.38423

BibTeX
@article{Zou2026,
  author = {Zou, Yuda and Zhang, Zijian and Xu, Yongchao},
  title = {Decoupling What to Count and Where to See for Referring Expression Counting},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {16},
  pages = {14113--14121},
  doi = {10.1609/aaai.v40i16.38423}
}
Kostenlos starten →