FastInstShadow: A Simple Query-Based Model for Instance Shadow Detection
This work addresses instance shadow detection, a domain-specific computer vision task, with incremental improvements in accuracy and speed for applications requiring real-time processing.
The paper tackles instance shadow detection by introducing FastInstShadow, a query-based model that directly associates shadows and objects during detection, achieving state-of-the-art performance on the SOBA dataset with improved accuracy over prior methods like SSISv2 and enabling real-time processing for moderate-resolution images.
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.