Event-STVG:面向長篇事件推理的時空影片定位基準
Event-STVG: Spatio-Temporal Video Grounding for Long-Form Event Reasoning
Event-STVG 提出面向多模態模型的長篇事件推理基準,要求模型在長影片中辨別重複事件,並進行多目標時空定位。基準涵蓋足球及另外七個領域,共標註 4,016 個事件;在 10 分鐘時長下,表現最強的模型 m-vIoU 只有 5.4%。研究亦觀察到模型難以區分同類重複事件、進行空間定位,以及同時定位所有所需角色。
Published on Oct 15
Authors:
,
,
,
,
,
,
,
,
,
,
,
Abstract
Event-STVG introduces a challenging benchmark requiring multimodal models to perform long-form event reasoning, repeated-event disambiguation, and multi-target spatio-temporal grounding across extended video contexts.
Spatio-Temporal Video Grounding (STVG) aims to jointly localize a target in space and time given a natural-language query. Despite continued advances, existing STVG benchmarks remain largely centered on localizing targets through instance-specific descriptions, leaving them largely decoupled from long-form reasoning increasingly emphasized in multimodal large language models (MLLMs). In this paper, we introduce Event-STVG, a task that shifts STVG from locating what is described toward event-level reasoning by unifying long-form context, repeated-event reasoning, event localization, role inference, and multi-target grounding. To facilitate research on Event-STVG, we construct a large-scale benchmark comprising 4,016 annotated event occurrences across soccer and seven additional domains. Experiments with state-of-the-art MLLMs reveal substantial difficulty, with the strongest model achieving only 5.4% m-vIoU at the 10-minute horizon. Across domains, we consistently observe confusion among repeated same-type events, spatial grounding bottlenecks, and rare success in jointly grounding all required roles. Event-STVG establishes a new challenge for visual understanding in MLLMs, calling for future models that can jointly reason about complex events across space and time. Code and annotations are available at https://github.com/THU-KEG/EventSTVG
View arXiv page View PDF Add to collection
Get this paper in your agent:
hf papers read 2610.22444
Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2610.22444 in a model README.md to link it from this page.
Datasets citing this paper 1
zacharyz21/EventSTVG
Viewer • Updated about 7 hours ago • 3.52k • 1.39k • 3
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2610.22444 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co