arXiv:2503.07517cs.CV2025-03被引 1

提出一种基于查询的实例阴影检测方法,提升精度与实时性。

FastInstShadow: A Simple Query-Based Model for Instance Shadow Detection

  • 用双路径变压器解码器直接关联阴影与物体
  • 在SOBA数据集上各项指标均优于现有方法
  • 适合需要高精度实时阴影检测的应用场景

实例阴影检测旨在识别阴影与其对应物体的配对。现有方法先独立检测阴影和物体,再进行关联。本文提出FastInstShadow,采用基于查询的架构,通过双路径变压器解码器在检测过程中评估阴影与物体间的关系。在SOBA数据集上的实验表明,该方法在所有评价指标上均超越现有方法。相比当前最准确的SSISv2,本方法在中等分辨率图像上实现了更高的精度与实时处理能力。代码已开源:https://github.com/wlotkr/FastInstShadow。

原文摘要 · Abstract (English)

Instance shadow detection is the task of detecting pairs of shadows and objects, where existing methods first detect shadows and objects independently, then associate them. This paper introduces FastInstShadow, a method that enhances detection accuracy through a query-based architecture featuring an association transformer decoder with two dual-path transformer decoders to assess relationships between shadows and objects during detection. Experimental results using the SOBA dataset showed that the proposed method outperforms all existing methods across all criteria. This method makes real-time processing feasible for moderate-resolution images with better accuracy than SSISv2, the most accurate existing method. Our code is available at https://github.com/wlotkr/FastInstShadow.

阴影检测实例分割实时推理

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。