偷看你的限動...........
In my latest creative explorations, I have been pushing the boundaries of spatial and temporal media. This includes experimenting with Point Cloud models to bridge the gap between physical reality and digital abstraction, creating immersive 3D environments that carry a unique aesthetic.
I am also deeply invested in the most advanced generative AI technologies. I utilize LTX Video 2.0/3.0 and FLUX models to produce high-fidelity cinematic content. Furthermore, I specialize in training custom LoRA (Low-Rank Adaptation) models to cultivate consistent virtual personas and unique visual styles for digital fashion and character modeling.
Beyond static and linear media, I am passionate about real-time interaction. By integrating TouchDesigner, I create responsive installations where AI-generated visuals react to human presence and environmental data, transforming the audience's role from passive observers to active participants.
My academic research at NYCU's Department of Applied Arts serves as the foundation for these experiments, where I examine the "symbiosis" between human creativity and machine intelligence in professional workflows.
Image Gen: Flux_krea Kontext, FLUX2-Klein, FireRed-ImageEdit, Qwen-image
Video Gen: LTX-Video 2.3, Wan2.1, Seedance 2.0
Voice & Audio: Index-TTS2, Qwen3-TTS
Training & Workflow: AI-Toolkit (LoRA Training), ComfyUI
Interaction: TouchDesigner, Point Cloud Modeling