Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Exploring Interactive Color Palettes for Abstraction-Driven Exploratory Image Colorization

Xinyu Shi, Mingyu Liu, Ziqi Zhou, Ali Neshati, Ryan Rossi, Jian Zhao

TL;DR

Findings suggest that interactive visual abstractions encourage a non-linear exploration workflow and an open mindset during ideation, thus providing better creative affordance.

Abstract

Color design is essential in areas such as product, graphic, and fashion design. However, current tools like Photoshop, with their concrete-driven color manipulation approach, often stumble during early ideation, favoring polished end results over initial exploration. We introduced Mondrian as a test-bed for abstraction-driven approach using interactive color palettes for image colorization. Through a formative study with six design experts, we selected three design options for visual abstractions in color design and developed Mondrian where humans work with abstractions and AI manages the concrete aspects. We carried out a user study to understand the benefits and challenges of each abstraction format and compare the Mondrian with Photoshop. A survey involving 100 participants further examined the influence of each abstraction format on color composition perceptions. Findings suggest that interactive visual abstractions encourage a non-linear exploration workflow and an open mindset during ideation, thus providing better creative affordance.

Exploring Interactive Color Palettes for Abstraction-Driven Exploratory Image Colorization

TL;DR

Findings suggest that interactive visual abstractions encourage a non-linear exploration workflow and an open mindset during ideation, thus providing better creative affordance.

Abstract

Color design is essential in areas such as product, graphic, and fashion design. However, current tools like Photoshop, with their concrete-driven color manipulation approach, often stumble during early ideation, favoring polished end results over initial exploration. We introduced Mondrian as a test-bed for abstraction-driven approach using interactive color palettes for image colorization. Through a formative study with six design experts, we selected three design options for visual abstractions in color design and developed Mondrian where humans work with abstractions and AI manages the concrete aspects. We carried out a user study to understand the benefits and challenges of each abstraction format and compare the Mondrian with Photoshop. A survey involving 100 participants further examined the influence of each abstraction format on color composition perceptions. Findings suggest that interactive visual abstractions encourage a non-linear exploration workflow and an open mindset during ideation, thus providing better creative affordance.
Paper Structure (31 sections, 10 figures, 1 table)

This paper contains 31 sections, 10 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. Abstraction in Creative Activities
  4. Color Palette Representations
  5. Interactive Tools for Color Design
  6. Color Aesthetic and Emotions
  7. Overview of Methodology
  8. Formative Study
  9. Participants and Procedure
  10. Design Insights
  11. Mondrian
  12. System Overview
  13. User Interface
  14. Palette Extraction
  15. Image Recolorization
  16. ...and 16 more sections

Figures (10)

  • Figure 1: The system pipeline of Mondrian, showing the main modules: (a) Palette Extraction which extracts 1D, 1D+, and 2D palettes; (b) Palette Manipulation, where users can modify colors and adjust quantities and proportions (specific to 1D+ palettes); and (c) Image Recolorization, where the recolorization model uses the modified palette and original image to generate the recolored image.
  • Figure 2: The user interface of Mondrian, consisting of Result Preview (A), Palette Manipulation (B), and Bookmark Tracking (C) panels. Palette Manipulation offers interaction with three palette formats (B1): 1D uniform (1D), 1D proportional (1D+), and 2D spatial (2D) palettes, adjustable with a global slider (B2). The Bookmark Tracking (C) panel stores intermediate states for assisting creativity exploration.
  • Figure 3: The images used in the user study, where (a) landscape photo and (b) artwork are for Task 1, and (c) robot with a person and (d) astronaut on the moon are used for Task 2.
  • Figure 4: (a) Results of Creative Support Index (CSI) for Mondrian and Photoshop (the higher the better) for the factors of exploration, expressiveness, results worth effort, immersion, immersion, enjoyment, and collaboration. (b) Participant-rated weights of each factor of CSI.
  • Figure 5: Comparison of designs using 1D+ format (A and B) and 2D format (C). Image D is a zoomed-in view of C, highlighting the difficulties participants experienced with the 2D palette in the sky area, which lacks distinct regions.
  • ...and 5 more figures