Elemental Alchemist: A Generative Interface for Semantic Control of Particle Systems Across Dynamic Levels of Abstraction

2026-05-11

AI Summary

This work addresses the challenge that users face in accurately mapping high-level creative intents—such as “make the fire look angry”—to low-level parameters in particle system visual effects editing. To bridge this gap, we propose a generative interactive interface that introduces, for the first time, a semantic control mechanism with dynamically adjustable abstraction levels. By integrating context-aware tool generation and parameter-to-semantic mapping, our system automatically translates user goals into multi-layered, steerable control signals. Designed to support collaborative creation by both non-expert and expert users, the approach significantly enhances editing efficiency and artistic controllability, as demonstrated in a user study with 15 participants. This effectively narrows the longstanding divide between creative intention and technical parameterization in visual effects authoring.

Abstract

Editing particle-system visual effects (VFX) is vital for digital storytelling, but achieving controllable, art-directable results remains challenging due to their multi-dimensional nature. Given a large collection of parameters, users must find the ones relevant to their creative goals -- a task that requires a systematic understanding of the particle system and how parameters map to high-level intents, such as making a fire look angry. Elemental Alchemist is a generative interface that transforms user intent into contextualized controls for semantic editing of particle systems. The system introduces two components: a contextual brush palette that generates tools based on scene context, and a generative control panel that surfaces relevant technical parameters and abstracts them to generate mid-level semantic attributes and high-level conceptual controls. An evaluation with 10 novice and 5 expert VFX practitioners shows the system supported users in translating high-level creative goals into particle system parameters.

Problem

• particle systems
• semantic control
• visual effects
• creative intent
• parameter abstraction

Innovation

• generative interface
• semantic control
• particle systems
• contextual brush palette
• abstraction levels