Git for Sketches: An Intelligent Tracking System for Capturing Design Evolution

TL;DR

DIMES delivers an integrated design ideation platform that captures the non-linear evolution and cognitive intent behind product concept sketches by embedding a stroke-aware version control system (sGIT) into a web-based environment. It combines a six-stroke taxonomy, the AEGIS data-gathering pipeline, and Generative AI to produce AI-driven narrative summaries and photorealistic renders, linking creative action to cognitive rationale. The system is validated through a comparative study showing increased exploration breadth ($H_1$), richer documentation via AI summaries ($H_2$), improved knowledge transfer as measured by Neural Transparency ($H_3$), and higher user acceptance of AI-rendered concepts ($H_4$). Collectively, DIMES demonstrates a paradigm shift toward cognitive design support systems that preserve design histories and enhance pedagogy and education within industrial design workflows.

Abstract

During product conceptualization, capturing the non-linear history and cognitive intent is crucial. Traditional sketching tools often lose this context. We introduce DIMES (Design Idea Management and Evolution capture System), a web-based environment featuring sGIT (SketchGit), a custom visual version control architecture, and Generative AI. sGIT includes AEGIS, a module using hybrid Deep Learning and Machine Learning models to classify six stroke types. The system maps Git primitives to design actions, enabling implicit branching and multi-modal commits (stroke data + voice intent). In a comparative study, experts using DIMES demonstrated a 160% increase in breadth of concept exploration. Generative AI modules generated narrative summaries that enhanced knowledge transfer; novices achieved higher replication fidelity (Neural Transparency-based Cosine Similarity: 0.97 vs. 0.73) compared to manual summaries. AI-generated renderings also received higher user acceptance (Purchase Likelihood: 4.2 vs 3.1). This work demonstrates that intelligent version control bridges creative action and cognitive documentation, offering a new paradigm for design education.

Figures (25)

• Figure 1: A diagram illustrating the design process as a converging funnel. It illustrates the stages of Concept, Development, and Detail Design, with cycles of divergence, convergence, and iteration culminating in manufacture.
• Figure 2: A classification chart of different types of illustrations used in design, categorized into Explorative, Persuasive, Explanatory, and Prescriptive, with the Prescriptive category highlighted and associated with CAD.
• Figure 3: A comparative visual taxonomy illustrating the fundamental differences between four distinct modes of visual representation in industrial design practice: a. Doodling – rapid, low-fidelity ideation with minimal structure and high ambiguity; b. Artistic Painting – expressive, painterly rendering prioritizing aesthetic impact and emotional communication over precision; c. Technical Drawing – dimensioned orthographic projections and cross-sections adhering to engineering drawing standards for manufacturing clarity; and d. Product Concept Sketch – purposeful combination of expressive form exploration, annotated design intent, construction awareness, and visual hierarchy to communicate industrial design concepts effectively during the fuzzy front-end of product development.
• Figure 4: Diagram illustrating the parallel between Goel's Goel1995 Lateral and Vertical Transformations in design cognition and VCS Branching and Versioning in software development. The left side depicts the exploration of alternatives (Lateral) versus the refinement of a single concept (Vertical). The right side shows the corresponding VCS concepts of parallel development through Branching and sequential history through Versioning.
• Figure 5: The progression of a coffee maker design from a rough concept (Stage 1) to a final detailed sketch (Stage 4), including specific component details like the grinder mechanism and control panel layout.