针对驾驶场景中小物体的指代表达理解，提出新数据集与渐进式放大定位方法。
Referring Expression Comprehension for Small Objects
- 设计渐进迭代缩放适配器，让模型逐步放大聚焦小物体。
- 在10万对样本上测试，显著提升小物体定位准确率。
- 适合自动驾驶、视觉定位等需要精准识别微小目标的场景。
指代表达理解（REC）旨在定位由自然语言描述的目标物体。近年来视觉-语言学习的进展显著提升了REC任务的表现，但极小物体的定位仍是重大挑战，尤其在自动驾驶等真实应用中至关重要。为此，本文提出一个面向小物体的新型数据集和方法。首先，构建了小物体指代表达理解（SOREC）数据集，包含10万对驾驶场景中描述小物体的自然语言表达及其对应边界框。其次，提出渐进迭代缩放适配器（PIZA），一种参数高效微调的适配模块，使模型能够逐步放大并精确定位小物体。通过一系列实验，将PIZA应用于GroundingDINO，在SOREC数据集上实现显著性能提升。相关数据集、代码及预训练模型已公开于项目页面。
原文摘要 · Abstract (English)
Referring expression comprehension (REC) aims to localize the target object described by a natural language expression. Recent advances in vision-language learning have led to significant performance improvements in REC tasks. However, localizing extremely small objects remains a considerable challenge despite its importance in real-world applications such as autonomous driving. To address this issue, we introduce a novel dataset and method for REC targeting small objects. First, we present the small object REC (SOREC) dataset, which consists of 100,000 pairs of referring expressions and corresponding bounding boxes for small objects in driving scenarios. Second, we propose the progressive-iterative zooming adapter (PIZA), an adapter module for parameter-efficient fine-tuning that enables models to progressively zoom in and localize small objects. In a series of experiments, we apply PIZA to GroundingDINO and demonstrate a significant improvement in accuracy on the SOREC dataset. Our dataset, codes and pre-trained models are publicly available on the project page.
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。