In this paper, we explore the unique modality of sketch for explainability, emphasising the profound impact of human strokes compared to conventional pixel-oriented studies. Beyond explanations of network behavior, we discern the genuine implications of explainability across diverse downstream sketch-related tasks. We propose a lightweight and portable explainability solution -- a seamless plugin that integrates effortlessly with any pre-trained model, eliminating the need for re-training. Demonstrating its adaptability, we present four applications: highly studied retrieval and generation, and completely novel assisted drawing and sketch adversarial attacks. The centrepiece to our solution is a stroke-level attribution map that takes different forms when linked with downstream tasks. By addressing the inherent non-differentiability of rasterisation, we enable explanations at both coarse stroke level (SLA) and partial stroke level (P-SLA), each with its advantages for specific downstream tasks.
@article{arxiv.2403.09480,
title = {What Sketch Explainability Really Means for Downstream Tasks},
author = {Hmrishav Bandyopadhyay and Pinaki Nath Chowdhury and Ayan Kumar Bhunia and Aneeshan Sain and Tao Xiang and Yi-Zhe Song},
journal= {arXiv preprint arXiv:2403.09480},
year = {2024}
}