Bridging Time and Space: Decoupled Spatio-Temporal Alignment for Video Grounding

TL;DR

This paper introduces Bridge-STG, a novel framework that decouples spatio-temporal localization in video grounding, effectively addressing entanglement and redundancy issues to achieve state-of-the-art results.

Contribution

Bridge-STG is the first end-to-end model to decouple temporal and spatial localization with semantic bridging and query-guided modules, improving performance on video grounding tasks.

Findings

Achieves state-of-the-art m_vIoU of 34.3 on VidSTG benchmark.

Significantly improves cross-task transfer in fine-grained video understanding.

Effectively reduces visual token redundancy in spatio-temporal grounding.

Abstract

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.