Saved in:
Bibliographic Details
Main Authors: Tu, Xuezhen, Wu, Jingyu, Kang, Fangyu, Nong, Qingpeng, Zhang, Kaijin, Niu, Chaoyue, Wu, Fan
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2604.08014
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918458575290368
author Tu, Xuezhen
Wu, Jingyu
Kang, Fangyu
Nong, Qingpeng
Zhang, Kaijin
Niu, Chaoyue
Wu, Fan
author_facet Tu, Xuezhen
Wu, Jingyu
Kang, Fangyu
Nong, Qingpeng
Zhang, Kaijin
Niu, Chaoyue
Wu, Fan
contents Spatio-Temporal Video Grounding requires jointly localizing target objects across both temporal and spatial dimensions based on natural language queries, posing fundamental challenges for existing Multimodal Large Language Models (MLLMs). We identify two core challenges: \textit{entangled spatio-temporal alignment}, arising from coupling two heterogeneous sub-tasks within the same autoregressive output space, and \textit{dual-domain visual token redundancy}, where target objects exhibit simultaneous temporal and spatial sparsity, rendering the overwhelming majority of visual tokens irrelevant to the grounding query. To address these, we propose \textbf{Bridge-STG}, an end-to-end framework that decouples temporal and spatial localization while maintaining semantic coherence. While decoupling is the natural solution to this entanglement, it risks creating a semantic gap between the temporal MLLM and the spatial decoder. Bridge-STG resolves this through two pivotal designs: the \textbf{Spatio-Temporal Semantic Bridging (STSB)} mechanism with Explicit Temporal Alignment (ETA) distills the MLLM's temporal reasoning context into enriched bridging queries as a robust semantic interface; and the \textbf{Query-Guided Spatial Localization (QGSL)} module leverages these queries to drive a purpose-built spatial decoder with multi-layer interactive queries and positive/negative frame sampling, jointly eliminating dual-domain visual token redundancy. Extensive experiments across multiple benchmarks demonstrate that Bridge-STG achieves state-of-the-art performance among MLLM-based methods. Bridge-STG improves average m\_vIoU from $26.4$ to $34.3$ on VidSTG and demonstrates strong cross-task transfer across various fine-grained video understanding tasks under a unified multi-task training regime.
format Preprint
id arxiv_https___arxiv_org_abs_2604_08014
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Bridging Time and Space: Decoupled Spatio-Temporal Alignment for Video Grounding
Tu, Xuezhen
Wu, Jingyu
Kang, Fangyu
Nong, Qingpeng
Zhang, Kaijin
Niu, Chaoyue
Wu, Fan
Computer Vision and Pattern Recognition
Spatio-Temporal Video Grounding requires jointly localizing target objects across both temporal and spatial dimensions based on natural language queries, posing fundamental challenges for existing Multimodal Large Language Models (MLLMs). We identify two core challenges: \textit{entangled spatio-temporal alignment}, arising from coupling two heterogeneous sub-tasks within the same autoregressive output space, and \textit{dual-domain visual token redundancy}, where target objects exhibit simultaneous temporal and spatial sparsity, rendering the overwhelming majority of visual tokens irrelevant to the grounding query. To address these, we propose \textbf{Bridge-STG}, an end-to-end framework that decouples temporal and spatial localization while maintaining semantic coherence. While decoupling is the natural solution to this entanglement, it risks creating a semantic gap between the temporal MLLM and the spatial decoder. Bridge-STG resolves this through two pivotal designs: the \textbf{Spatio-Temporal Semantic Bridging (STSB)} mechanism with Explicit Temporal Alignment (ETA) distills the MLLM's temporal reasoning context into enriched bridging queries as a robust semantic interface; and the \textbf{Query-Guided Spatial Localization (QGSL)} module leverages these queries to drive a purpose-built spatial decoder with multi-layer interactive queries and positive/negative frame sampling, jointly eliminating dual-domain visual token redundancy. Extensive experiments across multiple benchmarks demonstrate that Bridge-STG achieves state-of-the-art performance among MLLM-based methods. Bridge-STG improves average m\_vIoU from $26.4$ to $34.3$ on VidSTG and demonstrates strong cross-task transfer across various fine-grained video understanding tasks under a unified multi-task training regime.
title Bridging Time and Space: Decoupled Spatio-Temporal Alignment for Video Grounding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.08014