Metraipolis › archive › dc_30728fd6cf31
DeltaStreamer: Towards Human-Like Agentic Streaming Video Understanding — arXiv:2610.19849
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.19849
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T20:02:48.433Z
- SHA-256 of the archived text
9499d6c29ab07f2c0059cc52146d28284db73d31f1404f5a02d1244be3ed8bb7- Stable URL of this page
https://metraipolis.com/documento/dc_30728fd6cf31
How to cite
DeltaStreamer: Towards Human-Like Agentic Streaming Video Understanding — arXiv:2610.19849 — arXiv (en), https://arxiv.org/abs/2610.19849 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_30728fd6cf31 (Metraipolis archive dc_30728fd6cf31, sha256 9499d6c29ab07f2c0059cc52146d28284db73d31f1404f5a02d1244be3ed8bb7)
BibTeX
@misc{dc30728fd6cf31,
title = {DeltaStreamer: Towards Human-Like Agentic Streaming Video Understanding — arXiv:2610.19849},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.19849},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_30728fd6cf31; sha256 9499d6c29ab07f2c0059cc52146d28284db73d31f1404f5a02d1244be3ed8bb7},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19849. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
We propose DeltaStreamer, a novel agentic framework for streaming video understanding inspired by how humans observe, remember, and investigate. Streaming video understanding requires a system to retain evidence before knowing which details future queries will need. Existing approaches often obscure event transitions and omit relevant details when compressing video history, and cannot recover missing visual evidence. DeltaStreamer addresses these limitations by combining change-aware memory with query-driven planning and targeted evidence gathering with replay. Before a query arrives, it continuously observes the stream and explicitly records visual changes in structured memory, preserving how events unfold rather than only what is present. When a query arrives, a query-driven planner identifies the required evidence and guides memory retrieval and targeted replay. By revisiting historical frames, the agent can recover visual details omitted during observation and ground its answer in evidence beyond the retained descriptions. This connects continuous observation with active investigation tailored to each question. Experiments on OVO-Bench and StreamingBench demonstrate improvements over representative streaming video understanding methods, with overall scores of 66.07% and 59.53%, respectively. Project page: this https URL Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC). Pagina del lavoro: https://arxiv.org/abs/2610.19849 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-15T19:35:48.375Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).