VSI: Visual Subtitle Integration for Keyframe Selection to enhance Long Video Understanding

• URL: http://arxiv.org/abs/2508.06869v2
• Date: Sat, 06 Sep 2025 15:22:23 GMT
• Title: VSI: Visual Subtitle Integration for Keyframe Selection to enhance Long Video Understanding
• Authors: Jianxiang He, Meisheng Hong, Jungang Li, Yijie Xu, Ziyang Chen, Weiyu Guo, Hui Xiong,
• Abstract summary: Long video understanding presents a significant challenge to large language models (MLs)<n>VisualSubtitleation(VSI) integrates subtitles, semantic timestamps, and scene boundaries into a unified multimodal search process.<n>The proposed method captures the visual information of video frames as well as the complementary textual information through a dual-stream search mechanism.
• Score: 22.400847202448478
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Long video understanding presents a significant challenge to multimodal large language models (MLLMs) primarily due to the immense data scale. A critical and widely adopted strategy for making this task computationally tractable is keyframe retrieval, which seeks to identify a sparse set of video frames that are most salient to a given textual query. However, the efficacy of this approach is hindered by weak multimodal alignment between textual queries and visual content and fails to capture the complex temporal semantic information required for precise reasoning. To address this, we propose Visual-Subtitle Integeration(VSI), a multimodal keyframe search method that integrates subtitles, timestamps, and scene boundaries into a unified multimodal search process. The proposed method captures the visual information of video frames as well as the complementary textual information through a dual-stream search mechanism by Video Search Stream as well as Subtitle Match Stream, respectively, and improves the keyframe search accuracy through the interaction of the two search streams. Experimental results show that VSI achieve 40.00% key frame localization accuracy on the text-relevant subset of LongVideoBench and 68.48% accuracy on downstream long Video-QA tasks, surpassing competitive baselines by 20.35% and 15.79%, respectively. Furthermore, on the LongVideoBench, VSI achieved state-of-the-art(SOTA) in medium-to-long video-QA tasks, demonstrating the robustness and generalizability of the proposed multimodal search strategy.

Related papers

• IVCR-200K: A Large-Scale Multi-turn Dialogue Benchmark for Interactive Video Corpus Retrieval [36.33423199468626]
  Interactive Video Corpus Retrieval (IVCR) task enables multi-turn, conversational, and realistic interactions between the user and the retrieval system.<n> IVCR-200K is a high-quality, bilingual, multi-turn, conversational, and abstract semantic dataset that supports video retrieval and even moment retrieval.<n>We propose a comprehensive framework based on multi-modal large language models (MLLMs) to help users interact in several modes with more explainable solutions.
  arXiv  Detail & Related papers  (2025-12-01T06:12:59Z)
• CLaMR: Contextualized Late-Interaction for Multimodal Content Retrieval [70.9990850395981]
  We introduce CLaMR, a multimodal, late-interaction retriever that jointly indexes 4 modalities: video frames, transcribed speech, on-screen text, and metadata.<n>CLaMR is trained to enhance dynamic modality selection via two key innovations.
  arXiv  Detail & Related papers  (2025-06-06T15:02:30Z)
• Re-thinking Temporal Search for Long-Form Video Understanding [67.12801626407135]
  Current temporal search methods only achieve 2.1% temporal F1 score on the Longvideobench subset.<n>Inspired by visual search in images, we propose a lightweight temporal search framework, T* that reframes costly temporal search as spatial search.<n>Extensive experiments show that integrating T* with existing methods significantly improves SOTA long-form video understanding.
  arXiv  Detail & Related papers  (2025-04-03T04:03:10Z)
• Everything Can Be Described in Words: A Simple Unified Multi-Modal Framework with Semantic and Temporal Alignment [0.0]
  We propose UMaT, a framework that unifies visual and auditory inputs as structured text for large language models.<n>It significantly improves state-of-the-art Long Video Question Answering accuracy.
  arXiv  Detail & Related papers  (2025-03-12T05:28:24Z)
• HierarQ: Task-Aware Hierarchical Q-Former for Enhanced Video Understanding [14.464718780172582]
  We introduce HierarQ, a task-aware hierarchical Q-Former based framework that sequentially processes frames to bypass the need for frame sampling.<n>We introduce a lightweight two-stream language-guided feature modulator to incorporate task awareness in video understanding.<n>Extensive evaluations on 10 video benchmarks across video understanding, question answering, and captioning tasks demonstrate HierarQ's state-of-the-art performance.
  arXiv  Detail & Related papers  (2025-03-11T16:21:23Z)
• SALOVA: Segment-Augmented Long Video Assistant for Targeted Retrieval and Routing in Long-Form Video Analysis [52.050036778325094]
  We introduce SALOVA: Segment-Augmented Video Assistant, a novel video-LLM framework designed to enhance the comprehension of lengthy video content.<n>We present a high-quality collection of 87.8K long videos, each densely captioned at the segment level to enable models to capture scene continuity and maintain rich context.<n>Our framework mitigates the limitations of current video-LMMs by allowing for precise identification and retrieval of relevant video segments in response to queries.
  arXiv  Detail & Related papers  (2024-11-25T08:04:47Z)
• MultiVENT 2.0: A Massive Multilingual Benchmark for Event-Centric Video Retrieval [57.891157692501345]
  $textbfMultiVENT 2.0$ is a large-scale, multilingual event-centric video retrieval benchmark.<n>It features a collection of more than 218,000 news videos and 3,906 queries targeting specific world events.<n>Preliminary results show that state-of-the-art vision-language models struggle significantly with this task.
  arXiv  Detail & Related papers  (2024-10-15T13:56:34Z)
• Bridging Information Asymmetry in Text-video Retrieval: A Data-centric Approach [56.610806615527885]
  A key challenge in text-video retrieval (TVR) is the information asymmetry between video and text.<n>This paper introduces a data-centric framework to bridge this gap by enriching textual representations to better match the richness of video content.<n>We propose a query selection mechanism that identifies the most relevant and diverse queries, reducing computational cost while improving accuracy.
  arXiv  Detail & Related papers  (2024-08-14T01:24:09Z)
• Improving Video Corpus Moment Retrieval with Partial Relevance Enhancement [72.7576395034068]
  Video Corpus Moment Retrieval (VCMR) is a new video retrieval task aimed at retrieving a relevant moment from a large corpus of untrimmed videos using a text query. We argue that effectively capturing the partial relevance between the query and video is essential for the VCMR task. For video retrieval, we introduce a multi-modal collaborative video retriever, generating different query representations for the two modalities. For moment localization, we propose the focus-then-fuse moment localizer, utilizing modality-specific gates to capture essential content.
  arXiv  Detail & Related papers  (2024-02-21T07:16:06Z)
• A Challenging Multimodal Video Summary: Simultaneously Extracting and Generating Keyframe-Caption Pairs from Video [20.579167394855197]
  This paper proposes a practical multimodal video summarization task setting and dataset to train and evaluate the task. The target task involves summarizing a given video into a number ofcaption pairs and displaying them in a listable format to grasp the video content quickly. This task is useful as a practical application and presents a highly challenging problem worthy of study.
  arXiv  Detail & Related papers  (2023-12-04T02:17:14Z)
• Causal Video Summarizer for Video Exploration [74.27487067877047]
  Causal Video Summarizer (CVS) is proposed to capture the interactive information between the video and query. Based on the evaluation of the existing multi-modal video summarization dataset, experimental results show that the proposed approach is effective.
  arXiv  Detail & Related papers  (2023-07-04T22:52:16Z)
• Video Corpus Moment Retrieval with Contrastive Learning [56.249924768243375]
  Video corpus moment retrieval (VCMR) is to retrieve a temporal moment that semantically corresponds to a given text query. We propose a Retrieval and Localization Network with Contrastive Learning (ReLoCLNet) for VCMR. Experimental results show that ReLoCLNet encodes text and video separately for efficiency, its retrieval accuracy is comparable with baselines adopting cross-modal interaction learning.
  arXiv  Detail & Related papers  (2021-05-13T12:54:39Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.