GALAXY
China
Where do you locate yourself in relation to the systems you work with?
Where are you heading, and what is pulling you there?
How would you describe the space your practice is currently unfolding in?
Artist Statement
Description
ECDYSIS explores transformation as a continuous state rather than a finished result, using Chinese cultural symbolism as the foundation for digital fashion in the age of artificial intelligence. In traditional Chinese culture, the dragon is not a monster but a constructed symbol of order, authority, and collective imagination. The snake, by contrast, appears in ancient myths and folk traditions as a figure of shedding and instinct, closely linked to skin, bodily cycles, and sensory perception. It exists closer to the ground and closer to the body, representing change that originates from within. Rather than positioning the dragon and the snake as opposites, this work places them within a single evolving body. Their coexistence reflects a state of ecdysis, where identity is reshaped through shedding rather than replacement. This aligns with traditional Chinese understandings of transformation as gradual, cyclical, and rooted in nature. Floral and botanical forms extend this logic. In Chinese culture, plants symbolize time, resilience, and regeneration. Growth is depicted as an act of rupture: buds break soil, petals split open, and layers separate. Through virtual fashion and AIGC-driven design, these symbols become generative frameworks rather than decorative motifs. Fashion remains central, while AI enables cultural forms to evolve within a living digital system.
Process
This work began with a personal question: how can tradition continue to transform without being frozen as a symbol in the digital age? As artificial intelligence increasingly enters creative fields, I noticed that fashion is often treated as surface or spectacle, while cultural meaning becomes simplified or replicated. I was moved by the tension between rapid technological generation and the slow, embodied logic of traditional culture. I wanted to explore whether AI could participate in fashion without erasing history, and whether transformation could happen through evolution rather than replacement. Chinese cultural imagery offered a starting point. Symbols like the dragon, the snake, and botanical forms have always carried ideas of change, cycles, and regeneration. They do not represent fixed identities, but ongoing processes. This resonated with my understanding of both the body and technology. I followed the urge to treat fashion as a living system rather than a finished object. By working with virtual garments and AIGC-driven processes, I could observe how forms shed, reassemble, and evolve. This project is driven by a desire to create space where tradition, technology, and the body can continue to transform together, without losing their depth or complexity.
Tools
The process began with Stable Diffusion–based diffusion models, used to generate early garment sketches, line-based fashion studies, and visual storyboards. These text-to-image models functioned as a design laboratory, allowing me to explore silhouettes, surface logic, and material tension before any structural decisions were finalized. I also used diffusion outputs to iterate visual rhythm and shot composition for the film narrative. To expand visual language and perspective, Nanobanana and Midjourney plied to generate multiple viewpoints and alternative visual interpretations of each look, helping me test how the garments behave across different spatial and cinematic conditions. Selected designs were then translated into three-dimensional form using Tripo AI for AI-assisted 3D generation. These outputs were imported into Blender, where I refined garment structure, proportions, and body interaction, and designed camera paths, lighting, and movement. This stage focusing on volume, balance, and the relationship between garment and body. For motion and presentation, AI video generation and motion synthesis tools were used to simulate fabric movement, runway pacing, and cinematic transitions. Final sequences were assembled through AI-assisted editing systems, controlling continuity, and visual coherence. Throughout the process, AI operated as a generative engine, while all design decisions—selection, refinement, and fashion judgment—remained human-led.





