用事件实体提升图像检索效率，兼顾准确与速度。
Leveraging Lightweight Entity Extraction for Scalable Event-Based Image Retrieval
- 先提取关键实体做快速筛选，再用BEiT-3深度匹配
- 在OpenEvents上达0.559的MAP，显著优于旧方法
- 适合需要高效处理真实复杂查询的场景
从自然语言描述中检索图像是计算机视觉与自然语言处理交叉的核心任务，广泛应用于搜索引擎、媒体归档和数字内容管理。然而，真实场景下的图像文本检索仍面临查询模糊、语境依赖、语言多变及可扩展性要求高等挑战。本文提出一种轻量级两阶段检索流程，利用以事件为中心的实体抽取，融入真实描述中的时间与上下文信息。第一阶段基于关键实体使用BM25进行高效候选过滤，第二阶段采用BEiT-3模型捕捉深层多模态语义并重排序结果。在OpenEvents v1基准上，该方法达到0.559的均值平均精度，显著优于先前基线。结果表明，结合事件引导过滤与长文本视觉语言建模，在复杂真实场景下可实现精准高效的检索。代码已公开于https://github.com/PhamPhuHoa-23/Event-Based-Image-Retrieval。
原文摘要 · Abstract (English)
Retrieving images from natural language descriptions is a core task at the intersection of computer vision and natural language processing, with wide-ranging applications in search engines, media archiving, and digital content management. However, real-world image-text retrieval remains challenging due to vague or context-dependent queries, linguistic variability, and the need for scalable solutions. In this work, we propose a lightweight two-stage retrieval pipeline that leverages event-centric entity extraction to incorporate temporal and contextual signals from real-world captions. The first stage performs efficient candidate filtering using BM25 based on salient entities, while the second stage applies BEiT-3 models to capture deep multimodal semantics and rerank the results. Evaluated on the OpenEvents v1 benchmark, our method achieves a mean average precision of 0.559, substantially outperforming prior baselines. These results highlight the effectiveness of combining event-guided filtering with long-text vision-language modeling for accurate and efficient retrieval in complex, real-world scenarios. Our code is available at https://github.com/PhamPhuHoa-23/Event-Based-Image-Retrieval
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。