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.

我们在该研究中探索了草图的独特模态来增加可解释性，强调人类笔画相对于传统面向像素的研究的深远影响。我们提出了一个轻量级且便携的可解释性解决方案 —— 一种无缝插件，可以与任何预训练模型无缝集成，并消除了重新训练的需求。我们通过展示四种应用程序来演示其适应性：高度研究的检索和生成，以及全新的辅助绘图和草图对抗攻击。我们的解决方案的核心是一个笔画级归因图，在与下游任务相连接时采用不同形式。通过解决栅格化的固有非可微性，我们实现了对粗略笔画级（SLA）和部分笔画级（P-SLA）的解释，每个级别都具有特定下游任务的优势。

准确解释对下游任务的真正含义

What Sketch Explainability Really Means for Downstream Tasks

This paper presents a plugin that adds a representation of homogeneous and
heterogeneous, optically thick, translucent materials on the Blender 3D
modeling tool. The working principle of this plugin is based on a combination
of Genetic Algorithm (GA) and Singular Value Decomposition (SVD)-based
subsurface scattering method (GenSSS). The proposed plugin has been implemented
using Mitsuba renderer, which is an open source rendering software. The
proposed plugin has been validated on measured subsurface scattering data. It's
shown that the proposed plugin visualizes homogeneous and heterogeneous
subsurface scattering effects, accurately, compactly and computationally
efficiently.

本研究提出了一种插件，可以在 Blender 3D 建模工具上添加均质和异质的光学厚射线材料的表示。该插件的工作原理是基于遗传算法 (GA) 和奇异值分解 (SVD) 的次表面散射方法 (GenSSS) 的组合。所提出的插件利用开源渲染软件 Mitsuba 渲染器实现，并在已测量的次表面散射数据上进行了验证。结果表明，该插件可以准确、紧凑和计算效率高地可视化均质和异质次表面散射效应。