Em-Garde: A Propose-Match Framework for Proactive Streaming Video Understanding

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zheng, Yikai, Ding, Xin, Yang, Yifan, Jiang, Shiqi, Wu, Hao, Zhang, Qianxi, Wang, Weijun, Cao, Ting, Liu, Yunxin
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908901421613056
author Zheng, Yikai
Ding, Xin
Yang, Yifan
Jiang, Shiqi
Wu, Hao
Zhang, Qianxi
Wang, Weijun
Cao, Ting
Liu, Yunxin
author_facet Zheng, Yikai
Ding, Xin
Yang, Yifan
Jiang, Shiqi
Wu, Hao
Zhang, Qianxi
Wang, Weijun
Cao, Ting
Liu, Yunxin
contents Recent advances in Streaming Video Understanding has enabled a new interaction paradigm where models respond proactively to user queries. Current proactive VideoLLMs rely on per-frame triggering decision making, which suffers from an efficiency-accuracy dilemma. We propose Em-Garde, a novel framework that decouples semantic understanding from streaming perception. At query time, the Instruction-Guided Proposal Parser transforms user queries into structured, perceptually grounded visual proposals; during streaming, a Lightweight Proposal Matching Module performs efficient embedding-based matching to trigger responses. Experiments on StreamingBench and OVO-Bench demonstrate consistent improvements over prior models in proactive response accuracy and efficiency, validating an effective solution for proactive video understanding under strict computational constraints.
format Preprint
id arxiv_https___arxiv_org_abs_2603_19054
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Em-Garde: A Propose-Match Framework for Proactive Streaming Video Understanding
Zheng, Yikai
Ding, Xin
Yang, Yifan
Jiang, Shiqi
Wu, Hao
Zhang, Qianxi
Wang, Weijun
Cao, Ting
Liu, Yunxin
Computer Vision and Pattern Recognition
Artificial Intelligence
Recent advances in Streaming Video Understanding has enabled a new interaction paradigm where models respond proactively to user queries. Current proactive VideoLLMs rely on per-frame triggering decision making, which suffers from an efficiency-accuracy dilemma. We propose Em-Garde, a novel framework that decouples semantic understanding from streaming perception. At query time, the Instruction-Guided Proposal Parser transforms user queries into structured, perceptually grounded visual proposals; during streaming, a Lightweight Proposal Matching Module performs efficient embedding-based matching to trigger responses. Experiments on StreamingBench and OVO-Bench demonstrate consistent improvements over prior models in proactive response accuracy and efficiency, validating an effective solution for proactive video understanding under strict computational constraints.
title Em-Garde: A Propose-Match Framework for Proactive Streaming Video Understanding
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2603.19054