Explicit Temporal-Semantic Modeling for Dense Video Captioning via Context-Aware Cross-Modal Interaction
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (7) · 5341-5349
DOI: 10.1609/aaai.v40i7.37450 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Mingda Jia und 9 weiteren Autor:innen.
Abstract (Originalsprache)
Dense video captioning jointly localizes and captions salient events in untrimmed videos. Recent methods primarily focus on leveraging additional prior knowledge and advanced multi-task architectures to achieve competitive performance. However, these pipelines rely on implicit modeling that uses frame-level or fragmented video features, failing to capture the temporal coherence across event sequences and comprehensive semantics within visual contexts. To address this, we propose an explicit temporal-semantic modeling framework called Context-Aware Cross-Modal Interaction (CACMI), which leverages both latent temporal characteristics within videos and linguistic semantics from text corpus. Specifically, our model consists of two core components: Cross-modal Frame Aggregation aggregates relevant frames to extract temporally coherent, event-aligned textual features through cross-modal retrieval; and Context-aware Feature Enhancement utilizes query-guided attention to integrate visual dynamics with pseudo-event semantics. Extensive experiments on the ActivityNet Captions and YouCook2 datasets demonstrate that CACMI achieves the state-of-the-art performance on dense video captioning task.
Zitieren
APA
Jia, M., Meng, W., Fu, Z., Li, Y., Zeng, Q., Zhang, Y., Xin, J., Xu, R., Zhang, J., & Zhang, X. (2026). Explicit Temporal-Semantic Modeling for Dense Video Captioning via Context-Aware Cross-Modal Interaction. Proceedings of the AAAI Conference on Artificial Intelligence, 40(7), 5341-5349. https://doi.org/10.1609/aaai.v40i7.37450
Vancouver
Jia M, Meng W, Fu Z, Li Y, Zeng Q, Zhang Y, et al. Explicit Temporal-Semantic Modeling for Dense Video Captioning via Context-Aware Cross-Modal Interaction. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(7):5341-5349. doi:10.1609/aaai.v40i7.37450
BibTeX
@article{Jia2026,
author = {Jia, Mingda and Meng, Weiliang and Fu, Zenghuang and Li, Yiheng and Zeng, Qi and Zhang, Yifan and Xin, Ju and Xu, Rongtao and Zhang, Jiguang and Zhang, Xiaopeng},
title = {Explicit Temporal-Semantic Modeling for Dense Video Captioning via Context-Aware Cross-Modal Interaction},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {7},
pages = {5341--5349},
doi = {10.1609/aaai.v40i7.37450}
}
EUVIMED