Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Region-Aware Color Smudging

Ying Jiang, Pengfei Xu, Congyi Zhang, Hongbo Fu, Henry Lau, Wenping Wang

TL;DR

The paper tackles the challenge of generating natural shading in digital painting within a single layer by introducing SmartSmudge, a region-aware smudge tool that infers user intent from smudge paths with respect to color regions and employs a dynamic brush to preserve edges while enabling smooth gradients. It combines a region- and boundary-based region selection mechanism with a MeanShift-based color-region extraction and a dynamic masking approach to constrain smudging to target areas, formalized through the similarity score $R(s_t, \hat{\mathcal{T}}_t^*) = \alpha R_r + \beta R_b$ and thresholding with $\gamma$. A formative study identifies key challenges, which the method addresses via size-adaptive brushes, real-time region selection, and delayed-region correction, and a two-study user evaluation demonstrates improved efficiency, edge-preservation, and expressiveness compared with traditional smudging. The results suggest practical impact for artists and potential extensions to robotics, while acknowledging limitations in thin regions and the need for future global-factor integration and learning-based intent modeling.

Abstract

Color smudge operations from digital painting software enable users to create natural shading effects in high-fidelity paintings by interactively mixing colors. To precisely control results in traditional painting software, users tend to organize flat-filled color regions in multiple layers and smudge them to generate different color gradients. However, the requirement to carefully deal with regions makes the smudging process time-consuming and laborious, especially for non-professional users. This motivates us to investigate how to infer user-desired smudging effects when users smudge over regions in a single layer. To investigate improving color smudge performance, we first conduct a formative study. Following the findings of this study, we design SmartSmudge, a novel smudge tool that offers users dynamical smudge brushes and real-time region selection for easily generating natural and efficient shading effects. We demonstrate the efficiency and effectiveness of the proposed tool via a user study and quantitative analysis

Region-Aware Color Smudging

TL;DR

The paper tackles the challenge of generating natural shading in digital painting within a single layer by introducing SmartSmudge, a region-aware smudge tool that infers user intent from smudge paths with respect to color regions and employs a dynamic brush to preserve edges while enabling smooth gradients. It combines a region- and boundary-based region selection mechanism with a MeanShift-based color-region extraction and a dynamic masking approach to constrain smudging to target areas, formalized through the similarity score and thresholding with . A formative study identifies key challenges, which the method addresses via size-adaptive brushes, real-time region selection, and delayed-region correction, and a two-study user evaluation demonstrates improved efficiency, edge-preservation, and expressiveness compared with traditional smudging. The results suggest practical impact for artists and potential extensions to robotics, while acknowledging limitations in thin regions and the need for future global-factor integration and learning-based intent modeling.

Abstract

Color smudge operations from digital painting software enable users to create natural shading effects in high-fidelity paintings by interactively mixing colors. To precisely control results in traditional painting software, users tend to organize flat-filled color regions in multiple layers and smudge them to generate different color gradients. However, the requirement to carefully deal with regions makes the smudging process time-consuming and laborious, especially for non-professional users. This motivates us to investigate how to infer user-desired smudging effects when users smudge over regions in a single layer. To investigate improving color smudge performance, we first conduct a formative study. Following the findings of this study, we design SmartSmudge, a novel smudge tool that offers users dynamical smudge brushes and real-time region selection for easily generating natural and efficient shading effects. We demonstrate the efficiency and effectiveness of the proposed tool via a user study and quantitative analysis
Paper Structure (19 sections, 3 equations, 18 figures, 2 tables)

This paper contains 19 sections, 3 equations, 18 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Drawing Assistance Tools
  4. Image and Sketch Colorization
  5. Shading, Smoothing, and Smudging
  6. Formative Study
  7. User Study
  8. Observations
  9. Challenges
  10. Methodology
  11. Solutions
  12. Algorithm
  13. Evaluation
  14. Results and Discussion
  15. User Study
  16. ...and 4 more sections

Figures (18)

  • Figure 1: Left: a sketch drawn by a user. Middle: a flat-filled painting created by filling colors in the sketch. Right: the final painting created by blurring and smudging the flat-filled painting.
  • Figure 2: Left: the original painting and smudge strokes. Middle and Right: the respective shading effects after a blur operation and a smudge operation with the same strokes and the brush size is in 50 pixels.
  • Figure 3: Left: hard shading effects. Right: smooth shading effects.
  • Figure 4: Left: natural shading effects by our proposed tool SmartSmudge. Right: undesired shading effects caused by the traditional smudge tool.
  • Figure 5: The top and bottom rows show the paintings created by the same participants in the multi-layer mode and the single-layer mode, respectively. The first and fourth (from left to right) columns give the corresponding reference images.
  • ...and 13 more figures